id
large_string | submitter
large_string | authors
large_string | title
large_string | comments
large_string | journal-ref
large_string | doi
large_string | report-no
large_string | categories
large_string | license
large_string | abstract
large_string | versions
large list | update_date
timestamp[ms] | authors_parsed
large list | prompt
large_string | label
string | prob
float64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2203.16244 | Wei Lin | Wei Lin, Anna Kukleva, Kunyang Sun, Horst Possegger, Hilde Kuehne,
Horst Bischof | CycDA: Unsupervised Cycle Domain Adaptation from Image to Video | Accepted at ECCV2022. Supplementary included | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although action recognition has achieved impressive results over recent
years, both collection and annotation of video training data are still
time-consuming and cost intensive. Therefore, image-to-video adaptation has
been proposed to exploit labeling-free web image source for adapting on
unlabeled target videos. This poses two major challenges: (1) spatial domain
shift between web images and video frames; (2) modality gap between image and
video data. To address these challenges, we propose Cycle Domain Adaptation
(CycDA), a cycle-based approach for unsupervised image-to-video domain
adaptation by leveraging the joint spatial information in images and videos on
the one hand and, on the other hand, training an independent spatio-temporal
model to bridge the modality gap. We alternate between the spatial and
spatio-temporal learning with knowledge transfer between the two in each cycle.
We evaluate our approach on benchmark datasets for image-to-video as well as
for mixed-source domain adaptation achieving state-of-the-art results and
demonstrating the benefits of our cyclic adaptation. Code is available at
\url{https://github.com/wlin-at/CycDA}.
| [
{
"version": "v1",
"created": "Wed, 30 Mar 2022 12:22:26 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Jul 2022 12:05:53 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 11:19:19 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Lin",
"Wei",
""
],
[
"Kukleva",
"Anna",
""
],
[
"Sun",
"Kunyang",
""
],
[
"Possegger",
"Horst",
""
],
[
"Kuehne",
"Hilde",
""
],
[
"Bischof",
"Horst",
""
]
] | TITLE: CycDA: Unsupervised Cycle Domain Adaptation from Image to Video
ABSTRACT: Although action recognition has achieved impressive results over recent
years, both collection and annotation of video training data are still
time-consuming and cost intensive. Therefore, image-to-video adaptation has
been proposed to exploit labeling-free web image source for adapting on
unlabeled target videos. This poses two major challenges: (1) spatial domain
shift between web images and video frames; (2) modality gap between image and
video data. To address these challenges, we propose Cycle Domain Adaptation
(CycDA), a cycle-based approach for unsupervised image-to-video domain
adaptation by leveraging the joint spatial information in images and videos on
the one hand and, on the other hand, training an independent spatio-temporal
model to bridge the modality gap. We alternate between the spatial and
spatio-temporal learning with knowledge transfer between the two in each cycle.
We evaluate our approach on benchmark datasets for image-to-video as well as
for mixed-source domain adaptation achieving state-of-the-art results and
demonstrating the benefits of our cyclic adaptation. Code is available at
\url{https://github.com/wlin-at/CycDA}.
| no_new_dataset | 0.951459 |
2204.12037 | Yang Liu | Yang Liu, Yushen Wei, Hong Yan, Guanbin Li, Liang Lin | Causal Reasoning Meets Visual Representation Learning: A Prospective
Study | 35 pages, 14 figures. This work has been accepted by Machine
Intelligence Research. The arxiv version is kept updating by adding more
novel methods, datasets and insights. The official video interpretation of
this paper can be referred at https://youtu.be/2lfNaTkcTHI | null | null | null | cs.CV cs.AI cs.LG cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual representation learning is ubiquitous in various real-world
applications, including visual comprehension, video understanding, multi-modal
analysis, human-computer interaction, and urban computing. Due to the emergence
of huge amounts of multi-modal heterogeneous spatial/temporal/spatial-temporal
data in big data era, the lack of interpretability, robustness, and
out-of-distribution generalization are becoming the challenges of the existing
visual models. The majority of the existing methods tend to fit the original
data/variable distributions and ignore the essential causal relations behind
the multi-modal knowledge, which lacks unified guidance and analysis about why
modern visual representation learning methods easily collapse into data bias
and have limited generalization and cognitive abilities. Inspired by the strong
inference ability of human-level agents, recent years have therefore witnessed
great effort in developing causal reasoning paradigms to realize robust
representation and model learning with good cognitive ability. In this paper,
we conduct a comprehensive review of existing causal reasoning methods for
visual representation learning, covering fundamental theories, models, and
datasets. The limitations of current methods and datasets are also discussed.
Moreover, we propose some prospective challenges, opportunities, and future
research directions for benchmarking causal reasoning algorithms in visual
representation learning. This paper aims to provide a comprehensive overview of
this emerging field, attract attention, encourage discussions, bring to the
forefront the urgency of developing novel causal reasoning methods, publicly
available benchmarks, and consensus-building standards for reliable visual
representation learning and related real-world applications more efficiently.
| [
{
"version": "v1",
"created": "Tue, 26 Apr 2022 02:22:28 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Apr 2022 05:30:42 GMT"
},
{
"version": "v3",
"created": "Thu, 5 May 2022 03:39:04 GMT"
},
{
"version": "v4",
"created": "Mon, 9 May 2022 09:19:10 GMT"
},
{
"version": "v5",
"created": "Fri, 1 Jul 2022 08:35:07 GMT"
},
{
"version": "v6",
"created": "Tue, 2 Aug 2022 08:31:17 GMT"
},
{
"version": "v7",
"created": "Thu, 18 Aug 2022 08:16:26 GMT"
},
{
"version": "v8",
"created": "Wed, 22 Mar 2023 02:41:58 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Liu",
"Yang",
""
],
[
"Wei",
"Yushen",
""
],
[
"Yan",
"Hong",
""
],
[
"Li",
"Guanbin",
""
],
[
"Lin",
"Liang",
""
]
] | TITLE: Causal Reasoning Meets Visual Representation Learning: A Prospective
Study
ABSTRACT: Visual representation learning is ubiquitous in various real-world
applications, including visual comprehension, video understanding, multi-modal
analysis, human-computer interaction, and urban computing. Due to the emergence
of huge amounts of multi-modal heterogeneous spatial/temporal/spatial-temporal
data in big data era, the lack of interpretability, robustness, and
out-of-distribution generalization are becoming the challenges of the existing
visual models. The majority of the existing methods tend to fit the original
data/variable distributions and ignore the essential causal relations behind
the multi-modal knowledge, which lacks unified guidance and analysis about why
modern visual representation learning methods easily collapse into data bias
and have limited generalization and cognitive abilities. Inspired by the strong
inference ability of human-level agents, recent years have therefore witnessed
great effort in developing causal reasoning paradigms to realize robust
representation and model learning with good cognitive ability. In this paper,
we conduct a comprehensive review of existing causal reasoning methods for
visual representation learning, covering fundamental theories, models, and
datasets. The limitations of current methods and datasets are also discussed.
Moreover, we propose some prospective challenges, opportunities, and future
research directions for benchmarking causal reasoning algorithms in visual
representation learning. This paper aims to provide a comprehensive overview of
this emerging field, attract attention, encourage discussions, bring to the
forefront the urgency of developing novel causal reasoning methods, publicly
available benchmarks, and consensus-building standards for reliable visual
representation learning and related real-world applications more efficiently.
| no_new_dataset | 0.946547 |
2205.00363 | Fangyu Liu | Fangyu Liu, Guy Emerson, Nigel Collier | Visual Spatial Reasoning | TACL camera-ready version; code and data available at
https://github.com/cambridgeltl/visual-spatial-reasoning | null | null | null | cs.CL cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Spatial relations are a basic part of human cognition. However, they are
expressed in natural language in a variety of ways, and previous work has
suggested that current vision-and-language models (VLMs) struggle to capture
relational information. In this paper, we present Visual Spatial Reasoning
(VSR), a dataset containing more than 10k natural text-image pairs with 66
types of spatial relations in English (such as: under, in front of, and
facing). While using a seemingly simple annotation format, we show how the
dataset includes challenging linguistic phenomena, such as varying reference
frames. We demonstrate a large gap between human and model performance: the
human ceiling is above 95%, while state-of-the-art models only achieve around
70%. We observe that VLMs' by-relation performances have little correlation
with the number of training examples and the tested models are in general
incapable of recognising relations concerning the orientations of objects.
| [
{
"version": "v1",
"created": "Sat, 30 Apr 2022 23:03:49 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Feb 2023 18:42:00 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 15:42:50 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Liu",
"Fangyu",
""
],
[
"Emerson",
"Guy",
""
],
[
"Collier",
"Nigel",
""
]
] | TITLE: Visual Spatial Reasoning
ABSTRACT: Spatial relations are a basic part of human cognition. However, they are
expressed in natural language in a variety of ways, and previous work has
suggested that current vision-and-language models (VLMs) struggle to capture
relational information. In this paper, we present Visual Spatial Reasoning
(VSR), a dataset containing more than 10k natural text-image pairs with 66
types of spatial relations in English (such as: under, in front of, and
facing). While using a seemingly simple annotation format, we show how the
dataset includes challenging linguistic phenomena, such as varying reference
frames. We demonstrate a large gap between human and model performance: the
human ceiling is above 95%, while state-of-the-art models only achieve around
70%. We observe that VLMs' by-relation performances have little correlation
with the number of training examples and the tested models are in general
incapable of recognising relations concerning the orientations of objects.
| new_dataset | 0.96225 |
2205.02900 | Hao Zhang | Neil Jethani, Aahlad Puli, Hao Zhang, Leonid Garber, Lior Jankelson,
Yindalon Aphinyanaphongs, and Rajesh Ranganath | New-Onset Diabetes Assessment Using Artificial Intelligence-Enhanced
Electrocardiography | 21 pages, 8 figures | null | null | null | cs.LG cs.AI cs.CY | http://creativecommons.org/licenses/by-sa/4.0/ | Undiagnosed diabetes is present in 21.4% of adults with diabetes. Diabetes
can remain asymptomatic and undetected due to limitations in screening rates.
To address this issue, questionnaires, such as the American Diabetes
Association (ADA) Risk test, have been recommended for use by physicians and
the public. Based on evidence that blood glucose concentration can affect
cardiac electrophysiology, we hypothesized that an artificial intelligence
(AI)-enhanced electrocardiogram (ECG) could identify adults with new-onset
diabetes. We trained a neural network to estimate HbA1c using a 12-lead ECG and
readily available demographics. We retrospectively assembled a dataset
comprised of patients with paired ECG and HbA1c data. The population of
patients who receive both an ECG and HbA1c may a biased sample of the complete
outpatient population, so we adjusted the importance placed on each patient to
generate a more representative pseudo-population. We found ECG-based assessment
outperforms the ADA Risk test, achieving a higher area under the curve (0.80
vs. 0.68) and positive predictive value (13% vs. 9%) -- 2.6 times the
prevalence of diabetes in the cohort. The AI-enhanced ECG significantly
outperforms electrophysiologist interpretation of the ECG, suggesting that the
task is beyond current clinical capabilities. Given the prevalence of ECGs in
clinics and via wearable devices, such a tool would make precise, automated
diabetes assessment widely accessible.
| [
{
"version": "v1",
"created": "Thu, 5 May 2022 19:39:32 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 04:57:56 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Jethani",
"Neil",
""
],
[
"Puli",
"Aahlad",
""
],
[
"Zhang",
"Hao",
""
],
[
"Garber",
"Leonid",
""
],
[
"Jankelson",
"Lior",
""
],
[
"Aphinyanaphongs",
"Yindalon",
""
],
[
"Ranganath",
"Rajesh",
""
]
] | TITLE: New-Onset Diabetes Assessment Using Artificial Intelligence-Enhanced
Electrocardiography
ABSTRACT: Undiagnosed diabetes is present in 21.4% of adults with diabetes. Diabetes
can remain asymptomatic and undetected due to limitations in screening rates.
To address this issue, questionnaires, such as the American Diabetes
Association (ADA) Risk test, have been recommended for use by physicians and
the public. Based on evidence that blood glucose concentration can affect
cardiac electrophysiology, we hypothesized that an artificial intelligence
(AI)-enhanced electrocardiogram (ECG) could identify adults with new-onset
diabetes. We trained a neural network to estimate HbA1c using a 12-lead ECG and
readily available demographics. We retrospectively assembled a dataset
comprised of patients with paired ECG and HbA1c data. The population of
patients who receive both an ECG and HbA1c may a biased sample of the complete
outpatient population, so we adjusted the importance placed on each patient to
generate a more representative pseudo-population. We found ECG-based assessment
outperforms the ADA Risk test, achieving a higher area under the curve (0.80
vs. 0.68) and positive predictive value (13% vs. 9%) -- 2.6 times the
prevalence of diabetes in the cohort. The AI-enhanced ECG significantly
outperforms electrophysiologist interpretation of the ECG, suggesting that the
task is beyond current clinical capabilities. Given the prevalence of ECGs in
clinics and via wearable devices, such a tool would make precise, automated
diabetes assessment widely accessible.
| new_dataset | 0.964321 |
2205.06217 | Johannes Jakob Meyer | Johannes Jakob Meyer, Marian Mularski, Elies Gil-Fuster, Antonio Anna
Mele, Francesco Arzani, Alissa Wilms, Jens Eisert | Exploiting symmetry in variational quantum machine learning | 25 pages, 15 figures, comments welcome | PRX Quantum 4, 010328 (2023) | 10.1103/PRXQuantum.4.010328 | null | quant-ph cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Variational quantum machine learning is an extensively studied application of
near-term quantum computers. The success of variational quantum learning models
crucially depends on finding a suitable parametrization of the model that
encodes an inductive bias relevant to the learning task. However, precious
little is known about guiding principles for the construction of suitable
parametrizations. In this work, we holistically explore when and how symmetries
of the learning problem can be exploited to construct quantum learning models
with outcomes invariant under the symmetry of the learning task. Building on
tools from representation theory, we show how a standard gateset can be
transformed into an equivariant gateset that respects the symmetries of the
problem at hand through a process of gate symmetrization. We benchmark the
proposed methods on two toy problems that feature a non-trivial symmetry and
observe a substantial increase in generalization performance. As our tools can
also be applied in a straightforward way to other variational problems with
symmetric structure, we show how equivariant gatesets can be used in
variational quantum eigensolvers.
| [
{
"version": "v1",
"created": "Thu, 12 May 2022 17:01:41 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Meyer",
"Johannes Jakob",
""
],
[
"Mularski",
"Marian",
""
],
[
"Gil-Fuster",
"Elies",
""
],
[
"Mele",
"Antonio Anna",
""
],
[
"Arzani",
"Francesco",
""
],
[
"Wilms",
"Alissa",
""
],
[
"Eisert",
"Jens",
""
]
] | TITLE: Exploiting symmetry in variational quantum machine learning
ABSTRACT: Variational quantum machine learning is an extensively studied application of
near-term quantum computers. The success of variational quantum learning models
crucially depends on finding a suitable parametrization of the model that
encodes an inductive bias relevant to the learning task. However, precious
little is known about guiding principles for the construction of suitable
parametrizations. In this work, we holistically explore when and how symmetries
of the learning problem can be exploited to construct quantum learning models
with outcomes invariant under the symmetry of the learning task. Building on
tools from representation theory, we show how a standard gateset can be
transformed into an equivariant gateset that respects the symmetries of the
problem at hand through a process of gate symmetrization. We benchmark the
proposed methods on two toy problems that feature a non-trivial symmetry and
observe a substantial increase in generalization performance. As our tools can
also be applied in a straightforward way to other variational problems with
symmetric structure, we show how equivariant gatesets can be used in
variational quantum eigensolvers.
| no_new_dataset | 0.942188 |
2205.10873 | Corentin Dancette | Corentin Dancette and Matthieu Cord | Dynamic Query Selection for Fast Visual Perceiver | Accepted at the Transformer for Vision workshop, CVPR 2022 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Transformers have been matching deep convolutional networks for vision
architectures in recent works. Most work is focused on getting the best results
on large-scale benchmarks, and scaling laws seem to be the most successful
strategy: bigger models, more data, and longer training result in higher
performance. However, the reduction of network complexity and inference time
remains under-explored. The Perceiver model offers a solution to this problem:
by first performing a Cross-attention with a fixed number Q of latent query
tokens, the complexity of the L-layers Transformer network that follows is
bounded by O(LQ^2). In this work, we explore how to make Perceivers even more
efficient, by reducing the number of queries Q during inference while limiting
the accuracy drop.
| [
{
"version": "v1",
"created": "Sun, 22 May 2022 17:23:51 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 10:53:32 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Dancette",
"Corentin",
""
],
[
"Cord",
"Matthieu",
""
]
] | TITLE: Dynamic Query Selection for Fast Visual Perceiver
ABSTRACT: Transformers have been matching deep convolutional networks for vision
architectures in recent works. Most work is focused on getting the best results
on large-scale benchmarks, and scaling laws seem to be the most successful
strategy: bigger models, more data, and longer training result in higher
performance. However, the reduction of network complexity and inference time
remains under-explored. The Perceiver model offers a solution to this problem:
by first performing a Cross-attention with a fixed number Q of latent query
tokens, the complexity of the L-layers Transformer network that follows is
bounded by O(LQ^2). In this work, we explore how to make Perceivers even more
efficient, by reducing the number of queries Q during inference while limiting
the accuracy drop.
| no_new_dataset | 0.945851 |
2206.04192 | Aleksandar Pavlovi\'c | Aleksandar Pavlovi\'c and Emanuel Sallinger | ExpressivE: A Spatio-Functional Embedding For Knowledge Graph Completion | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge graphs are inherently incomplete. Therefore substantial research
has been directed toward knowledge graph completion (KGC), i.e., predicting
missing triples from the information represented in the knowledge graph (KG).
KG embedding models (KGEs) have yielded promising results for KGC, yet any
current KGE is incapable of: (1) fully capturing vital inference patterns
(e.g., composition), (2) capturing prominent patterns jointly (e.g., hierarchy
and composition), and (3) providing an intuitive interpretation of captured
patterns. In this work, we propose ExpressivE, a fully expressive
spatio-functional KGE that solves all these challenges simultaneously.
ExpressivE embeds pairs of entities as points and relations as
hyper-parallelograms in the virtual triple space $\mathbb{R}^{2d}$. This model
design allows ExpressivE not only to capture a rich set of inference patterns
jointly but additionally to display any supported inference pattern through the
spatial relation of hyper-parallelograms, offering an intuitive and consistent
geometric interpretation of ExpressivE embeddings and their captured patterns.
Experimental results on standard KGC benchmarks reveal that ExpressivE is
competitive with state-of-the-art KGEs and even significantly outperforms them
on WN18RR.
| [
{
"version": "v1",
"created": "Wed, 8 Jun 2022 23:34:39 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 11:20:01 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Pavlović",
"Aleksandar",
""
],
[
"Sallinger",
"Emanuel",
""
]
] | TITLE: ExpressivE: A Spatio-Functional Embedding For Knowledge Graph Completion
ABSTRACT: Knowledge graphs are inherently incomplete. Therefore substantial research
has been directed toward knowledge graph completion (KGC), i.e., predicting
missing triples from the information represented in the knowledge graph (KG).
KG embedding models (KGEs) have yielded promising results for KGC, yet any
current KGE is incapable of: (1) fully capturing vital inference patterns
(e.g., composition), (2) capturing prominent patterns jointly (e.g., hierarchy
and composition), and (3) providing an intuitive interpretation of captured
patterns. In this work, we propose ExpressivE, a fully expressive
spatio-functional KGE that solves all these challenges simultaneously.
ExpressivE embeds pairs of entities as points and relations as
hyper-parallelograms in the virtual triple space $\mathbb{R}^{2d}$. This model
design allows ExpressivE not only to capture a rich set of inference patterns
jointly but additionally to display any supported inference pattern through the
spatial relation of hyper-parallelograms, offering an intuitive and consistent
geometric interpretation of ExpressivE embeddings and their captured patterns.
Experimental results on standard KGC benchmarks reveal that ExpressivE is
competitive with state-of-the-art KGEs and even significantly outperforms them
on WN18RR.
| no_new_dataset | 0.946101 |
2206.05149 | Jizhizi Li | Jizhizi Li, Jing Zhang, Dacheng Tao | Referring Image Matting | Accepted to CVPR2023. The dataset, code and models are available at
https://github.com/JizhiziLi/RIM | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Different from conventional image matting, which either requires user-defined
scribbles/trimap to extract a specific foreground object or directly extracts
all the foreground objects in the image indiscriminately, we introduce a new
task named Referring Image Matting (RIM) in this paper, which aims to extract
the meticulous alpha matte of the specific object that best matches the given
natural language description, thus enabling a more natural and simpler
instruction for image matting. First, we establish a large-scale challenging
dataset RefMatte by designing a comprehensive image composition and expression
generation engine to automatically produce high-quality images along with
diverse text attributes based on public datasets. RefMatte consists of 230
object categories, 47,500 images, 118,749 expression-region entities, and
474,996 expressions. Additionally, we construct a real-world test set with 100
high-resolution natural images and manually annotate complex phrases to
evaluate the out-of-domain generalization abilities of RIM methods.
Furthermore, we present a novel baseline method CLIPMat for RIM, including a
context-embedded prompt, a text-driven semantic pop-up, and a multi-level
details extractor. Extensive experiments on RefMatte in both keyword and
expression settings validate the superiority of CLIPMat over representative
methods. We hope this work could provide novel insights into image matting and
encourage more follow-up studies. The dataset, code and models are available at
https://github.com/JizhiziLi/RIM.
| [
{
"version": "v1",
"created": "Fri, 10 Jun 2022 14:44:43 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Nov 2022 13:48:06 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 03:47:41 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Li",
"Jizhizi",
""
],
[
"Zhang",
"Jing",
""
],
[
"Tao",
"Dacheng",
""
]
] | TITLE: Referring Image Matting
ABSTRACT: Different from conventional image matting, which either requires user-defined
scribbles/trimap to extract a specific foreground object or directly extracts
all the foreground objects in the image indiscriminately, we introduce a new
task named Referring Image Matting (RIM) in this paper, which aims to extract
the meticulous alpha matte of the specific object that best matches the given
natural language description, thus enabling a more natural and simpler
instruction for image matting. First, we establish a large-scale challenging
dataset RefMatte by designing a comprehensive image composition and expression
generation engine to automatically produce high-quality images along with
diverse text attributes based on public datasets. RefMatte consists of 230
object categories, 47,500 images, 118,749 expression-region entities, and
474,996 expressions. Additionally, we construct a real-world test set with 100
high-resolution natural images and manually annotate complex phrases to
evaluate the out-of-domain generalization abilities of RIM methods.
Furthermore, we present a novel baseline method CLIPMat for RIM, including a
context-embedded prompt, a text-driven semantic pop-up, and a multi-level
details extractor. Extensive experiments on RefMatte in both keyword and
expression settings validate the superiority of CLIPMat over representative
methods. We hope this work could provide novel insights into image matting and
encourage more follow-up studies. The dataset, code and models are available at
https://github.com/JizhiziLi/RIM.
| new_dataset | 0.912942 |
2206.10555 | Chen Yukang | Yukang Chen, Jianhui Liu, Xiangyu Zhang, Xiaojuan Qi, Jiaya Jia | LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs | In CVPR 2023. Code is at
https://github.com/dvlab-research/LargeKernel3D | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advance in 2D CNNs has revealed that large kernels are important.
However, when directly applying large convolutional kernels in 3D CNNs, severe
difficulties are met, where those successful module designs in 2D become
surprisingly ineffective on 3D networks, including the popular depth-wise
convolution. To address this vital challenge, we instead propose the
spatial-wise partition convolution and its large-kernel module. As a result, it
avoids the optimization and efficiency issues of naive 3D large kernels. Our
large-kernel 3D CNN network, LargeKernel3D, yields notable improvement in 3D
tasks of semantic segmentation and object detection. It achieves 73.9% mIoU on
the ScanNetv2 semantic segmentation and 72.8% NDS nuScenes object detection
benchmarks, ranking 1st on the nuScenes LIDAR leaderboard. The performance
further boosts to 74.2% NDS with a simple multi-modal fusion. In addition,
LargeKernel3D can be scaled to 17x17x17 kernel size on Waymo 3D object
detection. For the first time, we show that large kernels are feasible and
essential for 3D visual tasks.
| [
{
"version": "v1",
"created": "Tue, 21 Jun 2022 17:35:57 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 12:43:10 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Chen",
"Yukang",
""
],
[
"Liu",
"Jianhui",
""
],
[
"Zhang",
"Xiangyu",
""
],
[
"Qi",
"Xiaojuan",
""
],
[
"Jia",
"Jiaya",
""
]
] | TITLE: LargeKernel3D: Scaling up Kernels in 3D Sparse CNNs
ABSTRACT: Recent advance in 2D CNNs has revealed that large kernels are important.
However, when directly applying large convolutional kernels in 3D CNNs, severe
difficulties are met, where those successful module designs in 2D become
surprisingly ineffective on 3D networks, including the popular depth-wise
convolution. To address this vital challenge, we instead propose the
spatial-wise partition convolution and its large-kernel module. As a result, it
avoids the optimization and efficiency issues of naive 3D large kernels. Our
large-kernel 3D CNN network, LargeKernel3D, yields notable improvement in 3D
tasks of semantic segmentation and object detection. It achieves 73.9% mIoU on
the ScanNetv2 semantic segmentation and 72.8% NDS nuScenes object detection
benchmarks, ranking 1st on the nuScenes LIDAR leaderboard. The performance
further boosts to 74.2% NDS with a simple multi-modal fusion. In addition,
LargeKernel3D can be scaled to 17x17x17 kernel size on Waymo 3D object
detection. For the first time, we show that large kernels are feasible and
essential for 3D visual tasks.
| no_new_dataset | 0.948298 |
2206.15083 | Jingyi Zhang | Jingyi Zhang, Jiaxing Huang, Xiaoqin Zhang, Shijian Lu | UniDAformer: Unified Domain Adaptive Panoptic Segmentation Transformer
via Hierarchical Mask Calibration | Accepted to CVPR2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Domain adaptive panoptic segmentation aims to mitigate data annotation
challenge by leveraging off-the-shelf annotated data in one or multiple related
source domains. However, existing studies employ two separate networks for
instance segmentation and semantic segmentation which lead to excessive network
parameters as well as complicated and computationally intensive training and
inference processes. We design UniDAformer, a unified domain adaptive panoptic
segmentation transformer that is simple but can achieve domain adaptive
instance segmentation and semantic segmentation simultaneously within a single
network. UniDAformer introduces Hierarchical Mask Calibration (HMC) that
rectifies inaccurate predictions at the level of regions, superpixels and
pixels via online self-training on the fly. It has three unique features: 1) it
enables unified domain adaptive panoptic adaptation; 2) it mitigates false
predictions and improves domain adaptive panoptic segmentation effectively; 3)
it is end-to-end trainable with a much simpler training and inference pipeline.
Extensive experiments over multiple public benchmarks show that UniDAformer
achieves superior domain adaptive panoptic segmentation as compared with the
state-of-the-art.
| [
{
"version": "v1",
"created": "Thu, 30 Jun 2022 07:32:23 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 05:52:23 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhang",
"Jingyi",
""
],
[
"Huang",
"Jiaxing",
""
],
[
"Zhang",
"Xiaoqin",
""
],
[
"Lu",
"Shijian",
""
]
] | TITLE: UniDAformer: Unified Domain Adaptive Panoptic Segmentation Transformer
via Hierarchical Mask Calibration
ABSTRACT: Domain adaptive panoptic segmentation aims to mitigate data annotation
challenge by leveraging off-the-shelf annotated data in one or multiple related
source domains. However, existing studies employ two separate networks for
instance segmentation and semantic segmentation which lead to excessive network
parameters as well as complicated and computationally intensive training and
inference processes. We design UniDAformer, a unified domain adaptive panoptic
segmentation transformer that is simple but can achieve domain adaptive
instance segmentation and semantic segmentation simultaneously within a single
network. UniDAformer introduces Hierarchical Mask Calibration (HMC) that
rectifies inaccurate predictions at the level of regions, superpixels and
pixels via online self-training on the fly. It has three unique features: 1) it
enables unified domain adaptive panoptic adaptation; 2) it mitigates false
predictions and improves domain adaptive panoptic segmentation effectively; 3)
it is end-to-end trainable with a much simpler training and inference pipeline.
Extensive experiments over multiple public benchmarks show that UniDAformer
achieves superior domain adaptive panoptic segmentation as compared with the
state-of-the-art.
| no_new_dataset | 0.950549 |
2208.00690 | Jae Won Cho | Jae Won Cho, Dong-jin Kim, Hyeonggon Ryu, In So Kweon | Generative Bias for Robust Visual Question Answering | CVPR 2023 | null | null | null | cs.CV cs.AI cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | The task of Visual Question Answering (VQA) is known to be plagued by the
issue of VQA models exploiting biases within the dataset to make its final
prediction. Various previous ensemble based debiasing methods have been
proposed where an additional model is purposefully trained to be biased in
order to train a robust target model. However, these methods compute the bias
for a model simply from the label statistics of the training data or from
single modal branches. In this work, in order to better learn the bias a target
VQA model suffers from, we propose a generative method to train the bias model
directly from the target model, called GenB. In particular, GenB employs a
generative network to learn the bias in the target model through a combination
of the adversarial objective and knowledge distillation. We then debias our
target model with GenB as a bias model, and show through extensive experiments
the effects of our method on various VQA bias datasets including VQA-CP2,
VQA-CP1, GQA-OOD, and VQA-CE, and show state-of-the-art results with the LXMERT
architecture on VQA-CP2.
| [
{
"version": "v1",
"created": "Mon, 1 Aug 2022 08:58:02 GMT"
},
{
"version": "v2",
"created": "Tue, 2 Aug 2022 02:53:59 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 07:20:37 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Cho",
"Jae Won",
""
],
[
"Kim",
"Dong-jin",
""
],
[
"Ryu",
"Hyeonggon",
""
],
[
"Kweon",
"In So",
""
]
] | TITLE: Generative Bias for Robust Visual Question Answering
ABSTRACT: The task of Visual Question Answering (VQA) is known to be plagued by the
issue of VQA models exploiting biases within the dataset to make its final
prediction. Various previous ensemble based debiasing methods have been
proposed where an additional model is purposefully trained to be biased in
order to train a robust target model. However, these methods compute the bias
for a model simply from the label statistics of the training data or from
single modal branches. In this work, in order to better learn the bias a target
VQA model suffers from, we propose a generative method to train the bias model
directly from the target model, called GenB. In particular, GenB employs a
generative network to learn the bias in the target model through a combination
of the adversarial objective and knowledge distillation. We then debias our
target model with GenB as a bias model, and show through extensive experiments
the effects of our method on various VQA bias datasets including VQA-CP2,
VQA-CP1, GQA-OOD, and VQA-CE, and show state-of-the-art results with the LXMERT
architecture on VQA-CP2.
| no_new_dataset | 0.94428 |
2209.04187 | Dong Huang | Si-Guo Fang, Dong Huang, Xiao-Sha Cai, Chang-Dong Wang, Chaobo He,
Yong Tang | Efficient Multi-view Clustering via Unified and Discrete Bipartite Graph
Learning | Accepted by IEEE Transactions on Neural Networks and Learning Systems | null | 10.1109/TNNLS.2023.3261460 | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Although previous graph-based multi-view clustering algorithms have gained
significant progress, most of them are still faced with three limitations.
First, they often suffer from high computational complexity, which restricts
their applications in large-scale scenarios. Second, they usually perform graph
learning either at the single-view level or at the view-consensus level, but
often neglect the possibility of the joint learning of single-view and
consensus graphs. Third, many of them rely on the k-means for discretization of
the spectral embeddings, which lack the ability to directly learn the graph
with discrete cluster structure. In light of this, this paper presents an
efficient multi-view clustering approach via unified and discrete bipartite
graph learning (UDBGL). Specifically, the anchor-based subspace learning is
incorporated to learn the view-specific bipartite graphs from multiple views,
upon which the bipartite graph fusion is leveraged to learn a view-consensus
bipartite graph with adaptive weight learning. Further, the Laplacian rank
constraint is imposed to ensure that the fused bipartite graph has discrete
cluster structures (with a specific number of connected components). By
simultaneously formulating the view-specific bipartite graph learning, the
view-consensus bipartite graph learning, and the discrete cluster structure
learning into a unified objective function, an efficient minimization algorithm
is then designed to tackle this optimization problem and directly achieve a
discrete clustering solution without requiring additional partitioning, which
notably has linear time complexity in data size. Experiments on a variety of
multi-view datasets demonstrate the robustness and efficiency of our UDBGL
approach. The code is available at https://github.com/huangdonghere/UDBGL.
| [
{
"version": "v1",
"created": "Fri, 9 Sep 2022 08:51:01 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 15:19:17 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Fang",
"Si-Guo",
""
],
[
"Huang",
"Dong",
""
],
[
"Cai",
"Xiao-Sha",
""
],
[
"Wang",
"Chang-Dong",
""
],
[
"He",
"Chaobo",
""
],
[
"Tang",
"Yong",
""
]
] | TITLE: Efficient Multi-view Clustering via Unified and Discrete Bipartite Graph
Learning
ABSTRACT: Although previous graph-based multi-view clustering algorithms have gained
significant progress, most of them are still faced with three limitations.
First, they often suffer from high computational complexity, which restricts
their applications in large-scale scenarios. Second, they usually perform graph
learning either at the single-view level or at the view-consensus level, but
often neglect the possibility of the joint learning of single-view and
consensus graphs. Third, many of them rely on the k-means for discretization of
the spectral embeddings, which lack the ability to directly learn the graph
with discrete cluster structure. In light of this, this paper presents an
efficient multi-view clustering approach via unified and discrete bipartite
graph learning (UDBGL). Specifically, the anchor-based subspace learning is
incorporated to learn the view-specific bipartite graphs from multiple views,
upon which the bipartite graph fusion is leveraged to learn a view-consensus
bipartite graph with adaptive weight learning. Further, the Laplacian rank
constraint is imposed to ensure that the fused bipartite graph has discrete
cluster structures (with a specific number of connected components). By
simultaneously formulating the view-specific bipartite graph learning, the
view-consensus bipartite graph learning, and the discrete cluster structure
learning into a unified objective function, an efficient minimization algorithm
is then designed to tackle this optimization problem and directly achieve a
discrete clustering solution without requiring additional partitioning, which
notably has linear time complexity in data size. Experiments on a variety of
multi-view datasets demonstrate the robustness and efficiency of our UDBGL
approach. The code is available at https://github.com/huangdonghere/UDBGL.
| no_new_dataset | 0.948965 |
2209.09385 | Zixiang Zhou | Dongqiangzi Ye, Zixiang Zhou, Weijia Chen, Yufei Xie, Yu Wang, Panqu
Wang and Hassan Foroosh | LidarMultiNet: Towards a Unified Multi-Task Network for LiDAR Perception | Accepted to AAAI 2023 (Oral). Full-length paper extending our
previous technical report of the 1st place solution of the 2022 Waymo Open
Dataset 3D Semantic Segmentation challenge, including evaluations on 5 major
benchmarks. arXiv admin note: text overlap with arXiv:2206.11428 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | LiDAR-based 3D object detection, semantic segmentation, and panoptic
segmentation are usually implemented in specialized networks with distinctive
architectures that are difficult to adapt to each other. This paper presents
LidarMultiNet, a LiDAR-based multi-task network that unifies these three major
LiDAR perception tasks. Among its many benefits, a multi-task network can
reduce the overall cost by sharing weights and computation among multiple
tasks. However, it typically underperforms compared to independently combined
single-task models. The proposed LidarMultiNet aims to bridge the performance
gap between the multi-task network and multiple single-task networks. At the
core of LidarMultiNet is a strong 3D voxel-based encoder-decoder architecture
with a Global Context Pooling (GCP) module extracting global contextual
features from a LiDAR frame. Task-specific heads are added on top of the
network to perform the three LiDAR perception tasks. More tasks can be
implemented simply by adding new task-specific heads while introducing little
additional cost. A second stage is also proposed to refine the first-stage
segmentation and generate accurate panoptic segmentation results. LidarMultiNet
is extensively tested on both Waymo Open Dataset and nuScenes dataset,
demonstrating for the first time that major LiDAR perception tasks can be
unified in a single strong network that is trained end-to-end and achieves
state-of-the-art performance. Notably, LidarMultiNet reaches the official 1st
place in the Waymo Open Dataset 3D semantic segmentation challenge 2022 with
the highest mIoU and the best accuracy for most of the 22 classes on the test
set, using only LiDAR points as input. It also sets the new state-of-the-art
for a single model on the Waymo 3D object detection benchmark and three
nuScenes benchmarks.
| [
{
"version": "v1",
"created": "Mon, 19 Sep 2022 23:39:15 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 20:30:25 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ye",
"Dongqiangzi",
""
],
[
"Zhou",
"Zixiang",
""
],
[
"Chen",
"Weijia",
""
],
[
"Xie",
"Yufei",
""
],
[
"Wang",
"Yu",
""
],
[
"Wang",
"Panqu",
""
],
[
"Foroosh",
"Hassan",
""
]
] | TITLE: LidarMultiNet: Towards a Unified Multi-Task Network for LiDAR Perception
ABSTRACT: LiDAR-based 3D object detection, semantic segmentation, and panoptic
segmentation are usually implemented in specialized networks with distinctive
architectures that are difficult to adapt to each other. This paper presents
LidarMultiNet, a LiDAR-based multi-task network that unifies these three major
LiDAR perception tasks. Among its many benefits, a multi-task network can
reduce the overall cost by sharing weights and computation among multiple
tasks. However, it typically underperforms compared to independently combined
single-task models. The proposed LidarMultiNet aims to bridge the performance
gap between the multi-task network and multiple single-task networks. At the
core of LidarMultiNet is a strong 3D voxel-based encoder-decoder architecture
with a Global Context Pooling (GCP) module extracting global contextual
features from a LiDAR frame. Task-specific heads are added on top of the
network to perform the three LiDAR perception tasks. More tasks can be
implemented simply by adding new task-specific heads while introducing little
additional cost. A second stage is also proposed to refine the first-stage
segmentation and generate accurate panoptic segmentation results. LidarMultiNet
is extensively tested on both Waymo Open Dataset and nuScenes dataset,
demonstrating for the first time that major LiDAR perception tasks can be
unified in a single strong network that is trained end-to-end and achieves
state-of-the-art performance. Notably, LidarMultiNet reaches the official 1st
place in the Waymo Open Dataset 3D semantic segmentation challenge 2022 with
the highest mIoU and the best accuracy for most of the 22 classes on the test
set, using only LiDAR points as input. It also sets the new state-of-the-art
for a single model on the Waymo 3D object detection benchmark and three
nuScenes benchmarks.
| no_new_dataset | 0.945248 |
2209.11425 | Jintao Wang | Jintao Wang, Shiqi Gong, Qingqing Wu, Shaodan Ma | RIS-Aided MIMO Systems with Hardware Impairments: Robust Beamforming
Design and Analysis | 16 pages, 6 figures. Accepted by IEEE Transactions on Wireless
Communications | null | 10.1109/TWC.2023.3246990 | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reconfigurable intelligent surface (RIS) has been anticipated to be a novel
cost-effective technology to improve the performance of future wireless
systems. In this paper, we investigate a practical RIS-aided
multiple-input-multiple-output (MIMO) system in the presence of transceiver
hardware impairments, RIS phase noise and imperfect channel state information
(CSI). Joint design of the MIMO transceiver and RIS reflection matrix to
minimize the total average mean-square-error (MSE) of all data streams is
particularly considered. This joint design problem is non-convex and
challenging to solve due to the newly considered practical imperfections. To
tackle the issue, we first analyze the total average MSE by incorporating the
impacts of the above system imperfections. Then, in order to handle the tightly
coupled optimization variables and non-convex NP-hard constraints, an efficient
iterative algorithm based on alternating optimization (AO) framework is
proposed with guaranteed convergence, where each subproblem admits a
closed-form optimal solution by leveraging the majorization-minimization (MM)
technique. Moreover, via exploiting the special structure of the unit-modulus
constraints, we propose a modified Riemannian gradient ascent (RGA) algorithm
for the discrete RIS phase shift optimization. Furthermore, the optimality of
the proposed algorithm is validated under line-of-sight (LoS) channel
conditions, and the irreducible MSE floor effect induced by imperfections of
both hardware and CSI is also revealed in the high signal-to-noise ratio (SNR)
regime. Numerical results show the superior MSE performance of our proposed
algorithm over the adopted benchmark schemes, and demonstrate that increasing
the number of RIS elements is not always beneficial under the above system
imperfections.
| [
{
"version": "v1",
"created": "Fri, 23 Sep 2022 05:53:22 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 12:40:16 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wang",
"Jintao",
""
],
[
"Gong",
"Shiqi",
""
],
[
"Wu",
"Qingqing",
""
],
[
"Ma",
"Shaodan",
""
]
] | TITLE: RIS-Aided MIMO Systems with Hardware Impairments: Robust Beamforming
Design and Analysis
ABSTRACT: Reconfigurable intelligent surface (RIS) has been anticipated to be a novel
cost-effective technology to improve the performance of future wireless
systems. In this paper, we investigate a practical RIS-aided
multiple-input-multiple-output (MIMO) system in the presence of transceiver
hardware impairments, RIS phase noise and imperfect channel state information
(CSI). Joint design of the MIMO transceiver and RIS reflection matrix to
minimize the total average mean-square-error (MSE) of all data streams is
particularly considered. This joint design problem is non-convex and
challenging to solve due to the newly considered practical imperfections. To
tackle the issue, we first analyze the total average MSE by incorporating the
impacts of the above system imperfections. Then, in order to handle the tightly
coupled optimization variables and non-convex NP-hard constraints, an efficient
iterative algorithm based on alternating optimization (AO) framework is
proposed with guaranteed convergence, where each subproblem admits a
closed-form optimal solution by leveraging the majorization-minimization (MM)
technique. Moreover, via exploiting the special structure of the unit-modulus
constraints, we propose a modified Riemannian gradient ascent (RGA) algorithm
for the discrete RIS phase shift optimization. Furthermore, the optimality of
the proposed algorithm is validated under line-of-sight (LoS) channel
conditions, and the irreducible MSE floor effect induced by imperfections of
both hardware and CSI is also revealed in the high signal-to-noise ratio (SNR)
regime. Numerical results show the superior MSE performance of our proposed
algorithm over the adopted benchmark schemes, and demonstrate that increasing
the number of RIS elements is not always beneficial under the above system
imperfections.
| no_new_dataset | 0.942718 |
2210.05361 | Xiaole Tang | Xiaole Tang, Xile Zhao, Jun Liu, Jianli Wang, Yuchun Miao, Tieyong
Zeng | Uncertainty-Aware Unsupervised Image Deblurring with Deep Residual Prior | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Non-blind deblurring methods achieve decent performance under the accurate
blur kernel assumption. Since the kernel uncertainty (i.e. kernel error) is
inevitable in practice, semi-blind deblurring is suggested to handle it by
introducing the prior of the kernel (or induced) error. However, how to design
a suitable prior for the kernel (or induced) error remains challenging.
Hand-crafted prior, incorporating domain knowledge, generally performs well but
may lead to poor performance when kernel (or induced) error is complex.
Data-driven prior, which excessively depends on the diversity and abundance of
training data, is vulnerable to out-of-distribution blurs and images. To
address this challenge, we suggest a dataset-free deep residual prior for the
kernel induced error (termed as residual) expressed by a customized untrained
deep neural network, which allows us to flexibly adapt to different blurs and
images in real scenarios. By organically integrating the respective strengths
of deep priors and hand-crafted priors, we propose an unsupervised semi-blind
deblurring model which recovers the latent image from the blurry image and
inaccurate blur kernel. To tackle the formulated model, an efficient
alternating minimization algorithm is developed. Extensive experiments
demonstrate the favorable performance of the proposed method as compared to
data-driven and model-driven methods in terms of image quality and the
robustness to the kernel error.
| [
{
"version": "v1",
"created": "Sun, 9 Oct 2022 11:10:59 GMT"
},
{
"version": "v2",
"created": "Tue, 1 Nov 2022 06:54:23 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Mar 2023 08:22:29 GMT"
},
{
"version": "v4",
"created": "Sun, 19 Mar 2023 08:16:50 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Tang",
"Xiaole",
""
],
[
"Zhao",
"Xile",
""
],
[
"Liu",
"Jun",
""
],
[
"Wang",
"Jianli",
""
],
[
"Miao",
"Yuchun",
""
],
[
"Zeng",
"Tieyong",
""
]
] | TITLE: Uncertainty-Aware Unsupervised Image Deblurring with Deep Residual Prior
ABSTRACT: Non-blind deblurring methods achieve decent performance under the accurate
blur kernel assumption. Since the kernel uncertainty (i.e. kernel error) is
inevitable in practice, semi-blind deblurring is suggested to handle it by
introducing the prior of the kernel (or induced) error. However, how to design
a suitable prior for the kernel (or induced) error remains challenging.
Hand-crafted prior, incorporating domain knowledge, generally performs well but
may lead to poor performance when kernel (or induced) error is complex.
Data-driven prior, which excessively depends on the diversity and abundance of
training data, is vulnerable to out-of-distribution blurs and images. To
address this challenge, we suggest a dataset-free deep residual prior for the
kernel induced error (termed as residual) expressed by a customized untrained
deep neural network, which allows us to flexibly adapt to different blurs and
images in real scenarios. By organically integrating the respective strengths
of deep priors and hand-crafted priors, we propose an unsupervised semi-blind
deblurring model which recovers the latent image from the blurry image and
inaccurate blur kernel. To tackle the formulated model, an efficient
alternating minimization algorithm is developed. Extensive experiments
demonstrate the favorable performance of the proposed method as compared to
data-driven and model-driven methods in terms of image quality and the
robustness to the kernel error.
| no_new_dataset | 0.948155 |
2210.12239 | Matthew Dirks | Matthew Dirks, David Poole | Auto-Encoder Neural Network Incorporating X-Ray Fluorescence Fundamental
Parameters with Machine Learning | X-Ray Spectrometry 2023 | null | 10.1002/xrs.3340 | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We consider energy-dispersive X-ray Fluorescence (EDXRF) applications where
the fundamental parameters method is impractical such as when instrument
parameters are unavailable. For example, on a mining shovel or conveyor belt,
rocks are constantly moving (leading to varying angles of incidence and
distances) and there may be other factors not accounted for (like dust). Neural
networks do not require instrument and fundamental parameters but training
neural networks requires XRF spectra labelled with elemental composition, which
is often limited because of its expense. We develop a neural network model that
learns from limited labelled data and also benefits from domain knowledge by
learning to invert a forward model. The forward model uses transition energies
and probabilities of all elements and parameterized distributions to
approximate other fundamental and instrument parameters. We evaluate the model
and baseline models on a rock dataset from a lithium mineral exploration
project. Our model works particularly well for some low-Z elements (Li, Mg, Al,
and K) as well as some high-Z elements (Sn and Pb) despite these elements being
outside the suitable range for common spectrometers to directly measure, likely
owing to the ability of neural networks to learn correlations and non-linear
relationships.
| [
{
"version": "v1",
"created": "Fri, 21 Oct 2022 20:41:44 GMT"
},
{
"version": "v2",
"created": "Wed, 11 Jan 2023 18:57:44 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Mar 2023 19:33:46 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Dirks",
"Matthew",
""
],
[
"Poole",
"David",
""
]
] | TITLE: Auto-Encoder Neural Network Incorporating X-Ray Fluorescence Fundamental
Parameters with Machine Learning
ABSTRACT: We consider energy-dispersive X-ray Fluorescence (EDXRF) applications where
the fundamental parameters method is impractical such as when instrument
parameters are unavailable. For example, on a mining shovel or conveyor belt,
rocks are constantly moving (leading to varying angles of incidence and
distances) and there may be other factors not accounted for (like dust). Neural
networks do not require instrument and fundamental parameters but training
neural networks requires XRF spectra labelled with elemental composition, which
is often limited because of its expense. We develop a neural network model that
learns from limited labelled data and also benefits from domain knowledge by
learning to invert a forward model. The forward model uses transition energies
and probabilities of all elements and parameterized distributions to
approximate other fundamental and instrument parameters. We evaluate the model
and baseline models on a rock dataset from a lithium mineral exploration
project. Our model works particularly well for some low-Z elements (Li, Mg, Al,
and K) as well as some high-Z elements (Sn and Pb) despite these elements being
outside the suitable range for common spectrometers to directly measure, likely
owing to the ability of neural networks to learn correlations and non-linear
relationships.
| no_new_dataset | 0.950503 |
2211.06597 | Chao Fan | Chao Fan and Junhao Liang and Chuanfu Shen and Saihui Hou and Yongzhen
Huang and Shiqi Yu | OpenGait: Revisiting Gait Recognition Toward Better Practicality | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Gait recognition is one of the most critical long-distance identification
technologies and increasingly gains popularity in both research and industry
communities. Despite the significant progress made in indoor datasets, much
evidence shows that gait recognition techniques perform poorly in the wild.
More importantly, we also find that some conclusions drawn from indoor datasets
cannot be generalized to real applications. Therefore, the primary goal of this
paper is to present a comprehensive benchmark study for better practicality
rather than only a particular model for better performance. To this end, we
first develop a flexible and efficient gait recognition codebase named
OpenGait. Based on OpenGait, we deeply revisit the recent development of gait
recognition by re-conducting the ablative experiments. Encouragingly,we detect
some unperfect parts of certain prior woks, as well as new insights. Inspired
by these discoveries, we develop a structurally simple, empirically powerful,
and practically robust baseline model, GaitBase. Experimentally, we
comprehensively compare GaitBase with many current gait recognition methods on
multiple public datasets, and the results reflect that GaitBase achieves
significantly strong performance in most cases regardless of indoor or outdoor
situations. Code is available at https://github.com/ShiqiYu/OpenGait.
| [
{
"version": "v1",
"created": "Sat, 12 Nov 2022 07:24:29 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Nov 2022 08:18:19 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 07:02:46 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Fan",
"Chao",
""
],
[
"Liang",
"Junhao",
""
],
[
"Shen",
"Chuanfu",
""
],
[
"Hou",
"Saihui",
""
],
[
"Huang",
"Yongzhen",
""
],
[
"Yu",
"Shiqi",
""
]
] | TITLE: OpenGait: Revisiting Gait Recognition Toward Better Practicality
ABSTRACT: Gait recognition is one of the most critical long-distance identification
technologies and increasingly gains popularity in both research and industry
communities. Despite the significant progress made in indoor datasets, much
evidence shows that gait recognition techniques perform poorly in the wild.
More importantly, we also find that some conclusions drawn from indoor datasets
cannot be generalized to real applications. Therefore, the primary goal of this
paper is to present a comprehensive benchmark study for better practicality
rather than only a particular model for better performance. To this end, we
first develop a flexible and efficient gait recognition codebase named
OpenGait. Based on OpenGait, we deeply revisit the recent development of gait
recognition by re-conducting the ablative experiments. Encouragingly,we detect
some unperfect parts of certain prior woks, as well as new insights. Inspired
by these discoveries, we develop a structurally simple, empirically powerful,
and practically robust baseline model, GaitBase. Experimentally, we
comprehensively compare GaitBase with many current gait recognition methods on
multiple public datasets, and the results reflect that GaitBase achieves
significantly strong performance in most cases regardless of indoor or outdoor
situations. Code is available at https://github.com/ShiqiYu/OpenGait.
| no_new_dataset | 0.940298 |
2211.06627 | Zhixi Cai | Zhixi Cai, Shreya Ghosh, Kalin Stefanov, Abhinav Dhall, Jianfei Cai,
Hamid Rezatofighi, Reza Haffari, Munawar Hayat | MARLIN: Masked Autoencoder for facial video Representation LearnINg | CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This paper proposes a self-supervised approach to learn universal facial
representations from videos, that can transfer across a variety of facial
analysis tasks such as Facial Attribute Recognition (FAR), Facial Expression
Recognition (FER), DeepFake Detection (DFD), and Lip Synchronization (LS). Our
proposed framework, named MARLIN, is a facial video masked autoencoder, that
learns highly robust and generic facial embeddings from abundantly available
non-annotated web crawled facial videos. As a challenging auxiliary task,
MARLIN reconstructs the spatio-temporal details of the face from the densely
masked facial regions which mainly include eyes, nose, mouth, lips, and skin to
capture local and global aspects that in turn help in encoding generic and
transferable features. Through a variety of experiments on diverse downstream
tasks, we demonstrate MARLIN to be an excellent facial video encoder as well as
feature extractor, that performs consistently well across a variety of
downstream tasks including FAR (1.13% gain over supervised benchmark), FER
(2.64% gain over unsupervised benchmark), DFD (1.86% gain over unsupervised
benchmark), LS (29.36% gain for Frechet Inception Distance), and even in low
data regime. Our code and models are available at
https://github.com/ControlNet/MARLIN .
| [
{
"version": "v1",
"created": "Sat, 12 Nov 2022 10:29:05 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Dec 2022 03:47:40 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 09:32:26 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Cai",
"Zhixi",
""
],
[
"Ghosh",
"Shreya",
""
],
[
"Stefanov",
"Kalin",
""
],
[
"Dhall",
"Abhinav",
""
],
[
"Cai",
"Jianfei",
""
],
[
"Rezatofighi",
"Hamid",
""
],
[
"Haffari",
"Reza",
""
],
[
"Hayat",
"Munawar",
""
]
] | TITLE: MARLIN: Masked Autoencoder for facial video Representation LearnINg
ABSTRACT: This paper proposes a self-supervised approach to learn universal facial
representations from videos, that can transfer across a variety of facial
analysis tasks such as Facial Attribute Recognition (FAR), Facial Expression
Recognition (FER), DeepFake Detection (DFD), and Lip Synchronization (LS). Our
proposed framework, named MARLIN, is a facial video masked autoencoder, that
learns highly robust and generic facial embeddings from abundantly available
non-annotated web crawled facial videos. As a challenging auxiliary task,
MARLIN reconstructs the spatio-temporal details of the face from the densely
masked facial regions which mainly include eyes, nose, mouth, lips, and skin to
capture local and global aspects that in turn help in encoding generic and
transferable features. Through a variety of experiments on diverse downstream
tasks, we demonstrate MARLIN to be an excellent facial video encoder as well as
feature extractor, that performs consistently well across a variety of
downstream tasks including FAR (1.13% gain over supervised benchmark), FER
(2.64% gain over unsupervised benchmark), DFD (1.86% gain over unsupervised
benchmark), LS (29.36% gain for Frechet Inception Distance), and even in low
data regime. Our code and models are available at
https://github.com/ControlNet/MARLIN .
| no_new_dataset | 0.949435 |
2211.09590 | Yuxuan Zhou | Yuxuan Zhou, Zhi-Qi Cheng, Chao Li, Yanwen Fang, Yifeng Geng, Xuansong
Xie, Margret Keuper | Hypergraph Transformer for Skeleton-based Action Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Skeleton-based action recognition aims to recognize human actions given human
joint coordinates with skeletal interconnections. By defining a graph with
joints as vertices and their natural connections as edges, previous works
successfully adopted Graph Convolutional networks (GCNs) to model joint
co-occurrences and achieved superior performance. More recently, a limitation
of GCNs is identified, i.e., the topology is fixed after training. To relax
such a restriction, Self-Attention (SA) mechanism has been adopted to make the
topology of GCNs adaptive to the input, resulting in the state-of-the-art
hybrid models. Concurrently, attempts with plain Transformers have also been
made, but they still lag behind state-of-the-art GCN-based methods due to the
lack of structural prior. Unlike hybrid models, we propose a more elegant
solution to incorporate the bone connectivity into Transformer via a graph
distance embedding. Our embedding retains the information of skeletal structure
during training, whereas GCNs merely use it for initialization. More
importantly, we reveal an underlying issue of graph models in general, i.e.,
pairwise aggregation essentially ignores the high-order kinematic dependencies
between body joints. To fill this gap, we propose a new self-attention (SA)
mechanism on hypergraph, termed Hypergraph Self-Attention (HyperSA), to
incorporate intrinsic higher-order relations into the model. We name the
resulting model Hyperformer, and it beats state-of-the-art graph models w.r.t.
accuracy and efficiency on NTU RGB+D, NTU RGB+D 120, and Northwestern-UCLA
datasets.
| [
{
"version": "v1",
"created": "Thu, 17 Nov 2022 15:36:48 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Jan 2023 07:38:35 GMT"
},
{
"version": "v3",
"created": "Thu, 9 Mar 2023 06:59:47 GMT"
},
{
"version": "v4",
"created": "Mon, 13 Mar 2023 05:19:15 GMT"
},
{
"version": "v5",
"created": "Tue, 21 Mar 2023 17:34:34 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhou",
"Yuxuan",
""
],
[
"Cheng",
"Zhi-Qi",
""
],
[
"Li",
"Chao",
""
],
[
"Fang",
"Yanwen",
""
],
[
"Geng",
"Yifeng",
""
],
[
"Xie",
"Xuansong",
""
],
[
"Keuper",
"Margret",
""
]
] | TITLE: Hypergraph Transformer for Skeleton-based Action Recognition
ABSTRACT: Skeleton-based action recognition aims to recognize human actions given human
joint coordinates with skeletal interconnections. By defining a graph with
joints as vertices and their natural connections as edges, previous works
successfully adopted Graph Convolutional networks (GCNs) to model joint
co-occurrences and achieved superior performance. More recently, a limitation
of GCNs is identified, i.e., the topology is fixed after training. To relax
such a restriction, Self-Attention (SA) mechanism has been adopted to make the
topology of GCNs adaptive to the input, resulting in the state-of-the-art
hybrid models. Concurrently, attempts with plain Transformers have also been
made, but they still lag behind state-of-the-art GCN-based methods due to the
lack of structural prior. Unlike hybrid models, we propose a more elegant
solution to incorporate the bone connectivity into Transformer via a graph
distance embedding. Our embedding retains the information of skeletal structure
during training, whereas GCNs merely use it for initialization. More
importantly, we reveal an underlying issue of graph models in general, i.e.,
pairwise aggregation essentially ignores the high-order kinematic dependencies
between body joints. To fill this gap, we propose a new self-attention (SA)
mechanism on hypergraph, termed Hypergraph Self-Attention (HyperSA), to
incorporate intrinsic higher-order relations into the model. We name the
resulting model Hyperformer, and it beats state-of-the-art graph models w.r.t.
accuracy and efficiency on NTU RGB+D, NTU RGB+D 120, and Northwestern-UCLA
datasets.
| no_new_dataset | 0.953794 |
2211.12764 | Siteng Huang | Siteng Huang, Biao Gong, Yulin Pan, Jianwen Jiang, Yiliang Lv, Yuyuan
Li, Donglin Wang | VoP: Text-Video Co-operative Prompt Tuning for Cross-Modal Retrieval | Accepted by CVPR 2023 | null | null | null | cs.CV cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many recent studies leverage the pre-trained CLIP for text-video cross-modal
retrieval by tuning the backbone with additional heavy modules, which not only
brings huge computational burdens with much more parameters, but also leads to
the knowledge forgetting from upstream models. In this work, we propose the
VoP: Text-Video Co-operative Prompt Tuning for efficient tuning on the
text-video retrieval task. The proposed VoP is an end-to-end framework with
both video & text prompts introducing, which can be regarded as a powerful
baseline with only 0.1% trainable parameters. Further, based on the
spatio-temporal characteristics of videos, we develop three novel video prompt
mechanisms to improve the performance with different scales of trainable
parameters. The basic idea of the VoP enhancement is to model the frame
position, frame context, and layer function with specific trainable prompts,
respectively. Extensive experiments show that compared to full fine-tuning, the
enhanced VoP achieves a 1.4% average R@1 gain across five text-video retrieval
benchmarks with 6x less parameter overhead. The code will be available at
https://github.com/bighuang624/VoP.
| [
{
"version": "v1",
"created": "Wed, 23 Nov 2022 08:20:29 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Mar 2023 06:31:05 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 02:36:52 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Huang",
"Siteng",
""
],
[
"Gong",
"Biao",
""
],
[
"Pan",
"Yulin",
""
],
[
"Jiang",
"Jianwen",
""
],
[
"Lv",
"Yiliang",
""
],
[
"Li",
"Yuyuan",
""
],
[
"Wang",
"Donglin",
""
]
] | TITLE: VoP: Text-Video Co-operative Prompt Tuning for Cross-Modal Retrieval
ABSTRACT: Many recent studies leverage the pre-trained CLIP for text-video cross-modal
retrieval by tuning the backbone with additional heavy modules, which not only
brings huge computational burdens with much more parameters, but also leads to
the knowledge forgetting from upstream models. In this work, we propose the
VoP: Text-Video Co-operative Prompt Tuning for efficient tuning on the
text-video retrieval task. The proposed VoP is an end-to-end framework with
both video & text prompts introducing, which can be regarded as a powerful
baseline with only 0.1% trainable parameters. Further, based on the
spatio-temporal characteristics of videos, we develop three novel video prompt
mechanisms to improve the performance with different scales of trainable
parameters. The basic idea of the VoP enhancement is to model the frame
position, frame context, and layer function with specific trainable prompts,
respectively. Extensive experiments show that compared to full fine-tuning, the
enhanced VoP achieves a 1.4% average R@1 gain across five text-video retrieval
benchmarks with 6x less parameter overhead. The code will be available at
https://github.com/bighuang624/VoP.
| no_new_dataset | 0.950824 |
2211.15387 | Xidan Song | Xidan Song, Youcheng Sun, Mustafa A. Mustafa and Lucas Cordeiro | AIREPAIR: A Repair Platform for Neural Networks | null | null | 10.48550/arXiv.2211.15387 | null | cs.NE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present AIREPAIR, a platform for repairing neural networks. It features
the integration of existing network repair tools. Based on AIREPAIR, one can
run different repair methods on the same model, thus enabling the fair
comparison of different repair techniques. We evaluate AIREPAIR with three
state-of-the-art repair tools on popular deep-learning datasets and models. Our
evaluation confirms the utility of AIREPAIR, by comparing and analyzing the
results from different repair techniques. A demonstration is available at
https://youtu.be/UkKw5neeWhw.
| [
{
"version": "v1",
"created": "Thu, 24 Nov 2022 10:02:38 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 19:47:52 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Song",
"Xidan",
""
],
[
"Sun",
"Youcheng",
""
],
[
"Mustafa",
"Mustafa A.",
""
],
[
"Cordeiro",
"Lucas",
""
]
] | TITLE: AIREPAIR: A Repair Platform for Neural Networks
ABSTRACT: We present AIREPAIR, a platform for repairing neural networks. It features
the integration of existing network repair tools. Based on AIREPAIR, one can
run different repair methods on the same model, thus enabling the fair
comparison of different repair techniques. We evaluate AIREPAIR with three
state-of-the-art repair tools on popular deep-learning datasets and models. Our
evaluation confirms the utility of AIREPAIR, by comparing and analyzing the
results from different repair techniques. A demonstration is available at
https://youtu.be/UkKw5neeWhw.
| no_new_dataset | 0.940298 |
2212.00937 | Yanqing Shen | Yanqing Shen, Sanping Zhou, Jingwen Fu, Ruotong Wang, Shitao Chen, and
Nanning Zheng | StructVPR: Distill Structural Knowledge with Weighting Samples for
Visual Place Recognition | accepted by CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual place recognition (VPR) is usually considered as a specific image
retrieval problem. Limited by existing training frameworks, most deep
learning-based works cannot extract sufficiently stable global features from
RGB images and rely on a time-consuming re-ranking step to exploit spatial
structural information for better performance. In this paper, we propose
StructVPR, a novel training architecture for VPR, to enhance structural
knowledge in RGB global features and thus improve feature stability in a
constantly changing environment. Specifically, StructVPR uses segmentation
images as a more definitive source of structural knowledge input into a CNN
network and applies knowledge distillation to avoid online segmentation and
inference of seg-branch in testing. Considering that not all samples contain
high-quality and helpful knowledge, and some even hurt the performance of
distillation, we partition samples and weigh each sample's distillation loss to
enhance the expected knowledge precisely. Finally, StructVPR achieves
impressive performance on several benchmarks using only global retrieval and
even outperforms many two-stage approaches by a large margin. After adding
additional re-ranking, ours achieves state-of-the-art performance while
maintaining a low computational cost.
| [
{
"version": "v1",
"created": "Fri, 2 Dec 2022 02:52:01 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Dec 2022 01:12:30 GMT"
},
{
"version": "v3",
"created": "Fri, 10 Mar 2023 13:49:14 GMT"
},
{
"version": "v4",
"created": "Wed, 22 Mar 2023 00:30:39 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Shen",
"Yanqing",
""
],
[
"Zhou",
"Sanping",
""
],
[
"Fu",
"Jingwen",
""
],
[
"Wang",
"Ruotong",
""
],
[
"Chen",
"Shitao",
""
],
[
"Zheng",
"Nanning",
""
]
] | TITLE: StructVPR: Distill Structural Knowledge with Weighting Samples for
Visual Place Recognition
ABSTRACT: Visual place recognition (VPR) is usually considered as a specific image
retrieval problem. Limited by existing training frameworks, most deep
learning-based works cannot extract sufficiently stable global features from
RGB images and rely on a time-consuming re-ranking step to exploit spatial
structural information for better performance. In this paper, we propose
StructVPR, a novel training architecture for VPR, to enhance structural
knowledge in RGB global features and thus improve feature stability in a
constantly changing environment. Specifically, StructVPR uses segmentation
images as a more definitive source of structural knowledge input into a CNN
network and applies knowledge distillation to avoid online segmentation and
inference of seg-branch in testing. Considering that not all samples contain
high-quality and helpful knowledge, and some even hurt the performance of
distillation, we partition samples and weigh each sample's distillation loss to
enhance the expected knowledge precisely. Finally, StructVPR achieves
impressive performance on several benchmarks using only global retrieval and
even outperforms many two-stage approaches by a large margin. After adding
additional re-ranking, ours achieves state-of-the-art performance while
maintaining a low computational cost.
| no_new_dataset | 0.945851 |
2212.02809 | Shiqiang Du | Baokai Liu, Fengjie He, Shiqiang Du, Jiacheng Li, Wenjie Liu | An advanced YOLOv3 method for small object detection | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Small object detection has important application value in the fields of
autonomous driving and drone scene analysis. As one of the most advanced object
detection algorithms, YOLOv3 suffers some challenges when detecting small
objects, such as the problem of detection failure of small objects and occluded
objects. To solve these problems, an improved YOLOv3 algorithm for small object
detection is proposed. In the proposed method, the dilated convolutions mish
(DCM) module is introduced into the backbone network of YOLOv3 to improve the
feature expression ability by fusing the feature maps of different receptive
fields. In the neck network of YOLOv3, the convolutional block attention module
(CBAM) and multi-level fusion module are introduced to select the important
information for small object detection in the shallow network, suppress the
uncritical information, and use the fusion module to fuse the feature maps of
different scales, so as to improve the detection accuracy of the algorithm. In
addition, the Soft-NMS and Complete-IoU (CloU) strategies are applied to
candidate frame screening, which improves the accuracy of the algorithm for the
detection of occluded objects. The ablation experiment of the MS COCO2017
object detection task proves the effectiveness of several modules introduced in
this paper for small object detection. The experimental results on the MS
COCO2017, VOC2007, and VOC2012 datasets show that the Average Precision (AP) of
this method is 16.5%, 8.71%, and 9.68% higher than that of YOLOv3,
respectively.
| [
{
"version": "v1",
"created": "Tue, 6 Dec 2022 07:58:21 GMT"
},
{
"version": "v2",
"created": "Mon, 26 Dec 2022 13:38:37 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 04:08:48 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Liu",
"Baokai",
""
],
[
"He",
"Fengjie",
""
],
[
"Du",
"Shiqiang",
""
],
[
"Li",
"Jiacheng",
""
],
[
"Liu",
"Wenjie",
""
]
] | TITLE: An advanced YOLOv3 method for small object detection
ABSTRACT: Small object detection has important application value in the fields of
autonomous driving and drone scene analysis. As one of the most advanced object
detection algorithms, YOLOv3 suffers some challenges when detecting small
objects, such as the problem of detection failure of small objects and occluded
objects. To solve these problems, an improved YOLOv3 algorithm for small object
detection is proposed. In the proposed method, the dilated convolutions mish
(DCM) module is introduced into the backbone network of YOLOv3 to improve the
feature expression ability by fusing the feature maps of different receptive
fields. In the neck network of YOLOv3, the convolutional block attention module
(CBAM) and multi-level fusion module are introduced to select the important
information for small object detection in the shallow network, suppress the
uncritical information, and use the fusion module to fuse the feature maps of
different scales, so as to improve the detection accuracy of the algorithm. In
addition, the Soft-NMS and Complete-IoU (CloU) strategies are applied to
candidate frame screening, which improves the accuracy of the algorithm for the
detection of occluded objects. The ablation experiment of the MS COCO2017
object detection task proves the effectiveness of several modules introduced in
this paper for small object detection. The experimental results on the MS
COCO2017, VOC2007, and VOC2012 datasets show that the Average Precision (AP) of
this method is 16.5%, 8.71%, and 9.68% higher than that of YOLOv3,
respectively.
| no_new_dataset | 0.944944 |
2212.06726 | Matteo Ferrante | Matteo Ferrante, Tommaso Boccato, Nicola Toschi | Semantic Brain Decoding: from fMRI to conceptually similar image
reconstruction of visual stimuli | null | null | null | null | cs.CV q-bio.NC | http://creativecommons.org/licenses/by/4.0/ | Brain decoding is a field of computational neuroscience that uses measurable
brain activity to infer mental states or internal representations of perceptual
inputs. Therefore, we propose a novel approach to brain decoding that also
relies on semantic and contextual similarity. We employ an fMRI dataset of
natural image vision and create a deep learning decoding pipeline inspired by
the existence of both bottom-up and top-down processes in human vision. We
train a linear brain-to-feature model to map fMRI activity features to visual
stimuli features, assuming that the brain projects visual information onto a
space that is homeomorphic to the latent space represented by the last
convolutional layer of a pretrained convolutional neural network, which
typically collects a variety of semantic features that summarize and highlight
similarities and differences between concepts. These features are then
categorized in the latent space using a nearest-neighbor strategy, and the
results are used to condition a generative latent diffusion model to create
novel images. From fMRI data only, we produce reconstructions of visual stimuli
that match the original content very well on a semantic level, surpassing the
state of the art in previous literature. We evaluate our work and obtain good
results using a quantitative semantic metric (the Wu-Palmer similarity metric
over the WordNet lexicon, which had an average value of 0.57) and perform a
human evaluation experiment that resulted in correct evaluation, according to
the multiplicity of human criteria in evaluating image similarity, in over 80%
of the test set.
| [
{
"version": "v1",
"created": "Tue, 13 Dec 2022 16:54:08 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 11:17:29 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ferrante",
"Matteo",
""
],
[
"Boccato",
"Tommaso",
""
],
[
"Toschi",
"Nicola",
""
]
] | TITLE: Semantic Brain Decoding: from fMRI to conceptually similar image
reconstruction of visual stimuli
ABSTRACT: Brain decoding is a field of computational neuroscience that uses measurable
brain activity to infer mental states or internal representations of perceptual
inputs. Therefore, we propose a novel approach to brain decoding that also
relies on semantic and contextual similarity. We employ an fMRI dataset of
natural image vision and create a deep learning decoding pipeline inspired by
the existence of both bottom-up and top-down processes in human vision. We
train a linear brain-to-feature model to map fMRI activity features to visual
stimuli features, assuming that the brain projects visual information onto a
space that is homeomorphic to the latent space represented by the last
convolutional layer of a pretrained convolutional neural network, which
typically collects a variety of semantic features that summarize and highlight
similarities and differences between concepts. These features are then
categorized in the latent space using a nearest-neighbor strategy, and the
results are used to condition a generative latent diffusion model to create
novel images. From fMRI data only, we produce reconstructions of visual stimuli
that match the original content very well on a semantic level, surpassing the
state of the art in previous literature. We evaluate our work and obtain good
results using a quantitative semantic metric (the Wu-Palmer similarity metric
over the WordNet lexicon, which had an average value of 0.57) and perform a
human evaluation experiment that resulted in correct evaluation, according to
the multiplicity of human criteria in evaluating image similarity, in over 80%
of the test set.
| no_new_dataset | 0.948106 |
2212.07060 | Zhengwei Bai | Zhengwei Bai, Guoyuan Wu, Matthew J. Barth, Yongkang Liu, Emrah Akin
Sisbot, Kentaro Oguchi | VINet: Lightweight, Scalable, and Heterogeneous Cooperative Perception
for 3D Object Detection | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Utilizing the latest advances in Artificial Intelligence (AI), the computer
vision community is now witnessing an unprecedented evolution in all kinds of
perception tasks, particularly in object detection. Based on multiple spatially
separated perception nodes, Cooperative Perception (CP) has emerged to
significantly advance the perception of automated driving. However, current
cooperative object detection methods mainly focus on ego-vehicle efficiency
without considering the practical issues of system-wide costs. In this paper,
we introduce VINet, a unified deep learning-based CP network for scalable,
lightweight, and heterogeneous cooperative 3D object detection. VINet is the
first CP method designed from the standpoint of large-scale system-level
implementation and can be divided into three main phases: 1) Global
Pre-Processing and Lightweight Feature Extraction which prepare the data into
global style and extract features for cooperation in a lightweight manner; 2)
Two-Stream Fusion which fuses the features from scalable and heterogeneous
perception nodes; and 3) Central Feature Backbone and 3D Detection Head which
further process the fused features and generate cooperative detection results.
An open-source data experimental platform is designed and developed for CP
dataset acquisition and model evaluation. The experimental analysis shows that
VINet can reduce 84% system-level computational cost and 94% system-level
communication cost while improving the 3D detection accuracy.
| [
{
"version": "v1",
"created": "Wed, 14 Dec 2022 07:03:23 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 02:44:57 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Bai",
"Zhengwei",
""
],
[
"Wu",
"Guoyuan",
""
],
[
"Barth",
"Matthew J.",
""
],
[
"Liu",
"Yongkang",
""
],
[
"Sisbot",
"Emrah Akin",
""
],
[
"Oguchi",
"Kentaro",
""
]
] | TITLE: VINet: Lightweight, Scalable, and Heterogeneous Cooperative Perception
for 3D Object Detection
ABSTRACT: Utilizing the latest advances in Artificial Intelligence (AI), the computer
vision community is now witnessing an unprecedented evolution in all kinds of
perception tasks, particularly in object detection. Based on multiple spatially
separated perception nodes, Cooperative Perception (CP) has emerged to
significantly advance the perception of automated driving. However, current
cooperative object detection methods mainly focus on ego-vehicle efficiency
without considering the practical issues of system-wide costs. In this paper,
we introduce VINet, a unified deep learning-based CP network for scalable,
lightweight, and heterogeneous cooperative 3D object detection. VINet is the
first CP method designed from the standpoint of large-scale system-level
implementation and can be divided into three main phases: 1) Global
Pre-Processing and Lightweight Feature Extraction which prepare the data into
global style and extract features for cooperation in a lightweight manner; 2)
Two-Stream Fusion which fuses the features from scalable and heterogeneous
perception nodes; and 3) Central Feature Backbone and 3D Detection Head which
further process the fused features and generate cooperative detection results.
An open-source data experimental platform is designed and developed for CP
dataset acquisition and model evaluation. The experimental analysis shows that
VINet can reduce 84% system-level computational cost and 94% system-level
communication cost while improving the 3D detection accuracy.
| no_new_dataset | 0.946646 |
2212.08273 | Jinlong Li | Jinlong Li, Runsheng Xu, Xinyu Liu, Jin Ma, Zicheng Chi, Jiaqi Ma,
Hongkai Yu | Learning for Vehicle-to-Vehicle Cooperative Perception under Lossy
Communication | this paper was accepted by IEEE Transactions on Intelligent Vehicles | 2023 IEEE Transactions on Intelligent Vehicles | 10.1109/TIV.2023.3260040 | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Deep learning has been widely used in the perception (e.g., 3D object
detection) of intelligent vehicle driving. Due to the beneficial
Vehicle-to-Vehicle (V2V) communication, the deep learning based features from
other agents can be shared to the ego vehicle so as to improve the perception
of the ego vehicle. It is named as Cooperative Perception in the V2V research,
whose algorithms have been dramatically advanced recently. However, all the
existing cooperative perception algorithms assume the ideal V2V communication
without considering the possible lossy shared features because of the Lossy
Communication (LC) which is common in the complex real-world driving scenarios.
In this paper, we first study the side effect (e.g., detection performance
drop) by the lossy communication in the V2V Cooperative Perception, and then we
propose a novel intermediate LC-aware feature fusion method to relieve the side
effect of lossy communication by a LC-aware Repair Network (LCRN) and enhance
the interaction between the ego vehicle and other vehicles by a specially
designed V2V Attention Module (V2VAM) including intra-vehicle attention of ego
vehicle and uncertainty-aware inter-vehicle attention. The extensive experiment
on the public cooperative perception dataset OPV2V (based on digital-twin CARLA
simulator) demonstrates that the proposed method is quite effective for the
cooperative point cloud based 3D object detection under lossy V2V
communication.
| [
{
"version": "v1",
"created": "Fri, 16 Dec 2022 04:18:47 GMT"
},
{
"version": "v2",
"created": "Sat, 18 Mar 2023 21:38:20 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Li",
"Jinlong",
""
],
[
"Xu",
"Runsheng",
""
],
[
"Liu",
"Xinyu",
""
],
[
"Ma",
"Jin",
""
],
[
"Chi",
"Zicheng",
""
],
[
"Ma",
"Jiaqi",
""
],
[
"Yu",
"Hongkai",
""
]
] | TITLE: Learning for Vehicle-to-Vehicle Cooperative Perception under Lossy
Communication
ABSTRACT: Deep learning has been widely used in the perception (e.g., 3D object
detection) of intelligent vehicle driving. Due to the beneficial
Vehicle-to-Vehicle (V2V) communication, the deep learning based features from
other agents can be shared to the ego vehicle so as to improve the perception
of the ego vehicle. It is named as Cooperative Perception in the V2V research,
whose algorithms have been dramatically advanced recently. However, all the
existing cooperative perception algorithms assume the ideal V2V communication
without considering the possible lossy shared features because of the Lossy
Communication (LC) which is common in the complex real-world driving scenarios.
In this paper, we first study the side effect (e.g., detection performance
drop) by the lossy communication in the V2V Cooperative Perception, and then we
propose a novel intermediate LC-aware feature fusion method to relieve the side
effect of lossy communication by a LC-aware Repair Network (LCRN) and enhance
the interaction between the ego vehicle and other vehicles by a specially
designed V2V Attention Module (V2VAM) including intra-vehicle attention of ego
vehicle and uncertainty-aware inter-vehicle attention. The extensive experiment
on the public cooperative perception dataset OPV2V (based on digital-twin CARLA
simulator) demonstrates that the proposed method is quite effective for the
cooperative point cloud based 3D object detection under lossy V2V
communication.
| no_new_dataset | 0.947284 |
2301.09120 | Zining Chen | Zining Chen, Weiqiu Wang, Zhicheng Zhao, Aidong Men | Causality-based Dual-Contrastive Learning Framework for Domain
Generalization | Inadequate proof of the effectiveness of the method | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Domain Generalization (DG) is essentially a sub-branch of out-of-distribution
generalization, which trains models from multiple source domains and
generalizes to unseen target domains. Recently, some domain generalization
algorithms have emerged, but most of them were designed with non-transferable
complex architecture. Additionally, contrastive learning has become a promising
solution for simplicity and efficiency in DG. However, existing contrastive
learning neglected domain shifts that caused severe model confusions. In this
paper, we propose a Dual-Contrastive Learning (DCL) module on feature and
prototype contrast. Moreover, we design a novel Causal Fusion Attention (CFA)
module to fuse diverse views of a single image to attain prototype.
Furthermore, we introduce a Similarity-based Hard-pair Mining (SHM) strategy to
leverage information on diversity shift. Extensive experiments show that our
method outperforms state-of-the-art algorithms on three DG datasets. The
proposed algorithm can also serve as a plug-and-play module without usage of
domain labels.
| [
{
"version": "v1",
"created": "Sun, 22 Jan 2023 13:07:24 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 06:41:10 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Chen",
"Zining",
""
],
[
"Wang",
"Weiqiu",
""
],
[
"Zhao",
"Zhicheng",
""
],
[
"Men",
"Aidong",
""
]
] | TITLE: Causality-based Dual-Contrastive Learning Framework for Domain
Generalization
ABSTRACT: Domain Generalization (DG) is essentially a sub-branch of out-of-distribution
generalization, which trains models from multiple source domains and
generalizes to unseen target domains. Recently, some domain generalization
algorithms have emerged, but most of them were designed with non-transferable
complex architecture. Additionally, contrastive learning has become a promising
solution for simplicity and efficiency in DG. However, existing contrastive
learning neglected domain shifts that caused severe model confusions. In this
paper, we propose a Dual-Contrastive Learning (DCL) module on feature and
prototype contrast. Moreover, we design a novel Causal Fusion Attention (CFA)
module to fuse diverse views of a single image to attain prototype.
Furthermore, we introduce a Similarity-based Hard-pair Mining (SHM) strategy to
leverage information on diversity shift. Extensive experiments show that our
method outperforms state-of-the-art algorithms on three DG datasets. The
proposed algorithm can also serve as a plug-and-play module without usage of
domain labels.
| no_new_dataset | 0.945651 |
2301.12876 | Deyao Zhu | Deyao Zhu, Yuhui Wang, J\"urgen Schmidhuber, Mohamed Elhoseiny | Guiding Online Reinforcement Learning with Action-Free Offline
Pretraining | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Offline RL methods have been shown to reduce the need for environment
interaction by training agents using offline collected episodes. However, these
methods typically require action information to be logged during data
collection, which can be difficult or even impossible in some practical cases.
In this paper, we investigate the potential of using action-free offline
datasets to improve online reinforcement learning, name this problem
Reinforcement Learning with Action-Free Offline Pretraining (AFP-RL). We
introduce Action-Free Guide (AF-Guide), a method that guides online training by
extracting knowledge from action-free offline datasets. AF-Guide consists of an
Action-Free Decision Transformer (AFDT) implementing a variant of Upside-Down
Reinforcement Learning. It learns to plan the next states from the offline
dataset, and a Guided Soft Actor-Critic (Guided SAC) that learns online with
guidance from AFDT. Experimental results show that AF-Guide can improve sample
efficiency and performance in online training thanks to the knowledge from the
action-free offline dataset. Code is available at
https://github.com/Vision-CAIR/AF-Guide.
| [
{
"version": "v1",
"created": "Mon, 30 Jan 2023 13:30:56 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 09:52:25 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhu",
"Deyao",
""
],
[
"Wang",
"Yuhui",
""
],
[
"Schmidhuber",
"Jürgen",
""
],
[
"Elhoseiny",
"Mohamed",
""
]
] | TITLE: Guiding Online Reinforcement Learning with Action-Free Offline
Pretraining
ABSTRACT: Offline RL methods have been shown to reduce the need for environment
interaction by training agents using offline collected episodes. However, these
methods typically require action information to be logged during data
collection, which can be difficult or even impossible in some practical cases.
In this paper, we investigate the potential of using action-free offline
datasets to improve online reinforcement learning, name this problem
Reinforcement Learning with Action-Free Offline Pretraining (AFP-RL). We
introduce Action-Free Guide (AF-Guide), a method that guides online training by
extracting knowledge from action-free offline datasets. AF-Guide consists of an
Action-Free Decision Transformer (AFDT) implementing a variant of Upside-Down
Reinforcement Learning. It learns to plan the next states from the offline
dataset, and a Guided Soft Actor-Critic (Guided SAC) that learns online with
guidance from AFDT. Experimental results show that AF-Guide can improve sample
efficiency and performance in online training thanks to the knowledge from the
action-free offline dataset. Code is available at
https://github.com/Vision-CAIR/AF-Guide.
| no_new_dataset | 0.946794 |
2302.02005 | Aleksandra \'Ciprijanovi\'c | A. \'Ciprijanovi\'c, A. Lewis, K. Pedro, S. Madireddy, B. Nord, G. N.
Perdue, S. M. Wild | DeepAstroUDA: Semi-Supervised Universal Domain Adaptation for
Cross-Survey Galaxy Morphology Classification and Anomaly Detection | Accepted in Machine Learning Science and Technology (MLST); 24 pages,
14 figures | null | null | FERMILAB-PUB-23-034-CSAID | astro-ph.GA cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Artificial intelligence methods show great promise in increasing the quality
and speed of work with large astronomical datasets, but the high complexity of
these methods leads to the extraction of dataset-specific, non-robust features.
Therefore, such methods do not generalize well across multiple datasets. We
present a universal domain adaptation method, \textit{DeepAstroUDA}, as an
approach to overcome this challenge. This algorithm performs semi-supervised
domain adaptation and can be applied to datasets with different data
distributions and class overlaps. Non-overlapping classes can be present in any
of the two datasets (the labeled source domain, or the unlabeled target
domain), and the method can even be used in the presence of unknown classes. We
apply our method to three examples of galaxy morphology classification tasks of
different complexities ($3$-class and $10$-class problems), with anomaly
detection: 1) datasets created after different numbers of observing years from
a single survey (LSST mock data of $1$ and $10$ years of observations); 2) data
from different surveys (SDSS and DECaLS); and 3) data from observing fields
with different depths within one survey (wide field and Stripe 82 deep field of
SDSS). For the first time, we demonstrate the successful use of domain
adaptation between very discrepant observational datasets.
\textit{DeepAstroUDA} is capable of bridging the gap between two astronomical
surveys, increasing classification accuracy in both domains (up to $40\%$ on
the unlabeled data), and making model performance consistent across datasets.
Furthermore, our method also performs well as an anomaly detection algorithm
and successfully clusters unknown class samples even in the unlabeled target
dataset.
| [
{
"version": "v1",
"created": "Fri, 3 Feb 2023 21:20:58 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 17:03:51 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ćiprijanović",
"A.",
""
],
[
"Lewis",
"A.",
""
],
[
"Pedro",
"K.",
""
],
[
"Madireddy",
"S.",
""
],
[
"Nord",
"B.",
""
],
[
"Perdue",
"G. N.",
""
],
[
"Wild",
"S. M.",
""
]
] | TITLE: DeepAstroUDA: Semi-Supervised Universal Domain Adaptation for
Cross-Survey Galaxy Morphology Classification and Anomaly Detection
ABSTRACT: Artificial intelligence methods show great promise in increasing the quality
and speed of work with large astronomical datasets, but the high complexity of
these methods leads to the extraction of dataset-specific, non-robust features.
Therefore, such methods do not generalize well across multiple datasets. We
present a universal domain adaptation method, \textit{DeepAstroUDA}, as an
approach to overcome this challenge. This algorithm performs semi-supervised
domain adaptation and can be applied to datasets with different data
distributions and class overlaps. Non-overlapping classes can be present in any
of the two datasets (the labeled source domain, or the unlabeled target
domain), and the method can even be used in the presence of unknown classes. We
apply our method to three examples of galaxy morphology classification tasks of
different complexities ($3$-class and $10$-class problems), with anomaly
detection: 1) datasets created after different numbers of observing years from
a single survey (LSST mock data of $1$ and $10$ years of observations); 2) data
from different surveys (SDSS and DECaLS); and 3) data from observing fields
with different depths within one survey (wide field and Stripe 82 deep field of
SDSS). For the first time, we demonstrate the successful use of domain
adaptation between very discrepant observational datasets.
\textit{DeepAstroUDA} is capable of bridging the gap between two astronomical
surveys, increasing classification accuracy in both domains (up to $40\%$ on
the unlabeled data), and making model performance consistent across datasets.
Furthermore, our method also performs well as an anomaly detection algorithm
and successfully clusters unknown class samples even in the unlabeled target
dataset.
| no_new_dataset | 0.948917 |
2302.03262 | Naoto Yanai | Tomoya Matsumoto and Takayuki Miura and Naoto Yanai | Membership Inference Attacks against Diffusion Models | null | DLSP 2023 (co-located in IEEE S&P 2023) | null | null | cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Diffusion models have attracted attention in recent years as innovative
generative models. In this paper, we investigate whether a diffusion model is
resistant to a membership inference attack, which evaluates the privacy leakage
of a machine learning model. We primarily discuss the diffusion model from the
standpoints of comparison with a generative adversarial network (GAN) as
conventional models and hyperparameters unique to the diffusion model, i.e.,
time steps, sampling steps, and sampling variances. We conduct extensive
experiments with DDIM as a diffusion model and DCGAN as a GAN on the CelebA and
CIFAR-10 datasets in both white-box and black-box settings and then confirm if
the diffusion model is comparably resistant to a membership inference attack as
GAN. Next, we demonstrate that the impact of time steps is significant and
intermediate steps in a noise schedule are the most vulnerable to the attack.
We also found two key insights through further analysis. First, we identify
that DDIM is vulnerable to the attack for small sample sizes instead of
achieving a lower FID. Second, sampling steps in hyperparameters are important
for resistance to the attack, whereas the impact of sampling variances is quite
limited.
| [
{
"version": "v1",
"created": "Tue, 7 Feb 2023 05:20:20 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 14:31:42 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Matsumoto",
"Tomoya",
""
],
[
"Miura",
"Takayuki",
""
],
[
"Yanai",
"Naoto",
""
]
] | TITLE: Membership Inference Attacks against Diffusion Models
ABSTRACT: Diffusion models have attracted attention in recent years as innovative
generative models. In this paper, we investigate whether a diffusion model is
resistant to a membership inference attack, which evaluates the privacy leakage
of a machine learning model. We primarily discuss the diffusion model from the
standpoints of comparison with a generative adversarial network (GAN) as
conventional models and hyperparameters unique to the diffusion model, i.e.,
time steps, sampling steps, and sampling variances. We conduct extensive
experiments with DDIM as a diffusion model and DCGAN as a GAN on the CelebA and
CIFAR-10 datasets in both white-box and black-box settings and then confirm if
the diffusion model is comparably resistant to a membership inference attack as
GAN. Next, we demonstrate that the impact of time steps is significant and
intermediate steps in a noise schedule are the most vulnerable to the attack.
We also found two key insights through further analysis. First, we identify
that DDIM is vulnerable to the attack for small sample sizes instead of
achieving a lower FID. Second, sampling steps in hyperparameters are important
for resistance to the attack, whereas the impact of sampling variances is quite
limited.
| no_new_dataset | 0.946941 |
2302.12242 | Mengde Xu | Mengde Xu, Zheng Zhang, Fangyun Wei, Han Hu and Xiang Bai | Side Adapter Network for Open-Vocabulary Semantic Segmentation | CVPR2023 Highlight | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper presents a new framework for open-vocabulary semantic segmentation
with the pre-trained vision-language model, named Side Adapter Network (SAN).
Our approach models the semantic segmentation task as a region recognition
problem. A side network is attached to a frozen CLIP model with two branches:
one for predicting mask proposals, and the other for predicting attention bias
which is applied in the CLIP model to recognize the class of masks. This
decoupled design has the benefit CLIP in recognizing the class of mask
proposals. Since the attached side network can reuse CLIP features, it can be
very light. In addition, the entire network can be trained end-to-end, allowing
the side network to be adapted to the frozen CLIP model, which makes the
predicted mask proposals CLIP-aware. Our approach is fast, accurate, and only
adds a few additional trainable parameters. We evaluate our approach on
multiple semantic segmentation benchmarks. Our method significantly outperforms
other counterparts, with up to 18 times fewer trainable parameters and 19 times
faster inference speed. We hope our approach will serve as a solid baseline and
help ease future research in open-vocabulary semantic segmentation. The code
will be available at https://github.com/MendelXu/SAN.
| [
{
"version": "v1",
"created": "Thu, 23 Feb 2023 18:58:28 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 09:45:35 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Xu",
"Mengde",
""
],
[
"Zhang",
"Zheng",
""
],
[
"Wei",
"Fangyun",
""
],
[
"Hu",
"Han",
""
],
[
"Bai",
"Xiang",
""
]
] | TITLE: Side Adapter Network for Open-Vocabulary Semantic Segmentation
ABSTRACT: This paper presents a new framework for open-vocabulary semantic segmentation
with the pre-trained vision-language model, named Side Adapter Network (SAN).
Our approach models the semantic segmentation task as a region recognition
problem. A side network is attached to a frozen CLIP model with two branches:
one for predicting mask proposals, and the other for predicting attention bias
which is applied in the CLIP model to recognize the class of masks. This
decoupled design has the benefit CLIP in recognizing the class of mask
proposals. Since the attached side network can reuse CLIP features, it can be
very light. In addition, the entire network can be trained end-to-end, allowing
the side network to be adapted to the frozen CLIP model, which makes the
predicted mask proposals CLIP-aware. Our approach is fast, accurate, and only
adds a few additional trainable parameters. We evaluate our approach on
multiple semantic segmentation benchmarks. Our method significantly outperforms
other counterparts, with up to 18 times fewer trainable parameters and 19 times
faster inference speed. We hope our approach will serve as a solid baseline and
help ease future research in open-vocabulary semantic segmentation. The code
will be available at https://github.com/MendelXu/SAN.
| no_new_dataset | 0.95388 |
2303.08241 | Shyam Venkatasubramanian | Shyam Venkatasubramanian, Sandeep Gogineni, Bosung Kang, Ali Pezeshki,
Muralidhar Rangaswamy, Vahid Tarokh | Subspace Perturbation Analysis for Data-Driven Radar Target Localization | 6 pages, 3 figures. Submitted to 2023 IEEE Radar Conference
(RadarConf). Extension of arXiv:2209.02890 | null | null | null | cs.CV eess.SP | http://creativecommons.org/licenses/by/4.0/ | Recent works exploring data-driven approaches to classical problems in
adaptive radar have demonstrated promising results pertaining to the task of
radar target localization. Via the use of space-time adaptive processing (STAP)
techniques and convolutional neural networks, these data-driven approaches to
target localization have helped benchmark the performance of neural networks
for matched scenarios. However, the thorough bridging of these topics across
mismatched scenarios still remains an open problem. As such, in this work, we
augment our data-driven approach to radar target localization by performing a
subspace perturbation analysis, which allows us to benchmark the localization
accuracy of our proposed deep learning framework across mismatched scenarios.
To evaluate this framework, we generate comprehensive datasets by randomly
placing targets of variable strengths in mismatched constrained areas via
RFView, a high-fidelity, site-specific modeling and simulation tool. For the
radar returns from these constrained areas, we generate heatmap tensors in
range, azimuth, and elevation using the normalized adaptive matched filter
(NAMF) test statistic. We estimate target locations from these heatmap tensors
using a convolutional neural network, and demonstrate that the predictive
performance of our framework in the presence of mismatches can be
predetermined.
| [
{
"version": "v1",
"created": "Tue, 14 Mar 2023 21:22:26 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 21:24:53 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Venkatasubramanian",
"Shyam",
""
],
[
"Gogineni",
"Sandeep",
""
],
[
"Kang",
"Bosung",
""
],
[
"Pezeshki",
"Ali",
""
],
[
"Rangaswamy",
"Muralidhar",
""
],
[
"Tarokh",
"Vahid",
""
]
] | TITLE: Subspace Perturbation Analysis for Data-Driven Radar Target Localization
ABSTRACT: Recent works exploring data-driven approaches to classical problems in
adaptive radar have demonstrated promising results pertaining to the task of
radar target localization. Via the use of space-time adaptive processing (STAP)
techniques and convolutional neural networks, these data-driven approaches to
target localization have helped benchmark the performance of neural networks
for matched scenarios. However, the thorough bridging of these topics across
mismatched scenarios still remains an open problem. As such, in this work, we
augment our data-driven approach to radar target localization by performing a
subspace perturbation analysis, which allows us to benchmark the localization
accuracy of our proposed deep learning framework across mismatched scenarios.
To evaluate this framework, we generate comprehensive datasets by randomly
placing targets of variable strengths in mismatched constrained areas via
RFView, a high-fidelity, site-specific modeling and simulation tool. For the
radar returns from these constrained areas, we generate heatmap tensors in
range, azimuth, and elevation using the normalized adaptive matched filter
(NAMF) test statistic. We estimate target locations from these heatmap tensors
using a convolutional neural network, and demonstrate that the predictive
performance of our framework in the presence of mismatches can be
predetermined.
| no_new_dataset | 0.95096 |
2303.08996 | Aron Brenner | Aron Brenner, Rahman Khorramfar, Saurabh Amin | Learning Spatio-Temporal Aggregations for Large-Scale Capacity Expansion
Problems | null | null | null | null | cs.LG cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Effective investment planning decisions are crucial to ensure cyber-physical
infrastructures satisfy performance requirements over an extended time horizon.
Computing these decisions often requires solving Capacity Expansion Problems
(CEPs). In the context of regional-scale energy systems, these problems are
prohibitively expensive to solve due to large network sizes, heterogeneous node
characteristics, and a large number of operational periods. To maintain
tractability, traditional approaches aggregate network nodes and/or select a
set of representative time periods. Often, these reductions do not capture
supply-demand variations that crucially impact CEP costs and constraints,
leading to suboptimal decisions. Here, we propose a novel graph convolutional
autoencoder approach for spatio-temporal aggregation of a generic CEP with
heterogeneous nodes (CEPHN). Our architecture leverages graph pooling to
identify nodes with similar characteristics and minimizes a multi-objective
loss function. This loss function is tailored to induce desirable spatial and
temporal aggregations with regard to tractability and optimality. In
particular, the output of the graph pooling provides a spatial aggregation
while clustering the low-dimensional encoded representations yields a temporal
aggregation. We apply our approach to generation expansion planning of a
coupled 88-node power and natural gas system in New England. The resulting
aggregation leads to a simpler CEPHN with 6 nodes and a small set of
representative days selected from one year. We evaluate aggregation outcomes
over a range of hyperparameters governing the loss function and compare
resulting upper bounds on the original problem with those obtained using
benchmark methods. We show that our approach provides upper bounds that are 33%
(resp. 10%) lower those than obtained from benchmark spatial (resp. temporal)
aggregation approaches.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 00:00:52 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 02:42:51 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Brenner",
"Aron",
""
],
[
"Khorramfar",
"Rahman",
""
],
[
"Amin",
"Saurabh",
""
]
] | TITLE: Learning Spatio-Temporal Aggregations for Large-Scale Capacity Expansion
Problems
ABSTRACT: Effective investment planning decisions are crucial to ensure cyber-physical
infrastructures satisfy performance requirements over an extended time horizon.
Computing these decisions often requires solving Capacity Expansion Problems
(CEPs). In the context of regional-scale energy systems, these problems are
prohibitively expensive to solve due to large network sizes, heterogeneous node
characteristics, and a large number of operational periods. To maintain
tractability, traditional approaches aggregate network nodes and/or select a
set of representative time periods. Often, these reductions do not capture
supply-demand variations that crucially impact CEP costs and constraints,
leading to suboptimal decisions. Here, we propose a novel graph convolutional
autoencoder approach for spatio-temporal aggregation of a generic CEP with
heterogeneous nodes (CEPHN). Our architecture leverages graph pooling to
identify nodes with similar characteristics and minimizes a multi-objective
loss function. This loss function is tailored to induce desirable spatial and
temporal aggregations with regard to tractability and optimality. In
particular, the output of the graph pooling provides a spatial aggregation
while clustering the low-dimensional encoded representations yields a temporal
aggregation. We apply our approach to generation expansion planning of a
coupled 88-node power and natural gas system in New England. The resulting
aggregation leads to a simpler CEPHN with 6 nodes and a small set of
representative days selected from one year. We evaluate aggregation outcomes
over a range of hyperparameters governing the loss function and compare
resulting upper bounds on the original problem with those obtained using
benchmark methods. We show that our approach provides upper bounds that are 33%
(resp. 10%) lower those than obtained from benchmark spatial (resp. temporal)
aggregation approaches.
| no_new_dataset | 0.947769 |
2303.09540 | Daniel Simig | Amro Abbas, Kushal Tirumala, D\'aniel Simig, Surya Ganguli, Ari S.
Morcos | SemDeDup: Data-efficient learning at web-scale through semantic
deduplication | null | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Progress in machine learning has been driven in large part by massive
increases in data. However, large web-scale datasets such as LAION are largely
uncurated beyond searches for exact duplicates, potentially leaving much
redundancy. Here, we introduce SemDeDup, a method which leverages embeddings
from pre-trained models to identify and remove semantic duplicates: data pairs
which are semantically similar, but not exactly identical. Removing semantic
duplicates preserves performance and speeds up learning. Analyzing a subset of
LAION, we show that SemDeDup can remove 50% of the data with minimal
performance loss, effectively halving training time. Moreover, performance
increases out of distribution. Also, analyzing language models trained on C4, a
partially curated dataset, we show that SemDeDup improves over prior approaches
while providing efficiency gains. SemDeDup provides an example of how simple
ways of leveraging quality embeddings can be used to make models learn faster
with less data.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:53:24 GMT"
},
{
"version": "v2",
"created": "Sat, 18 Mar 2023 10:12:47 GMT"
},
{
"version": "v3",
"created": "Wed, 22 Mar 2023 17:22:35 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Abbas",
"Amro",
""
],
[
"Tirumala",
"Kushal",
""
],
[
"Simig",
"Dániel",
""
],
[
"Ganguli",
"Surya",
""
],
[
"Morcos",
"Ari S.",
""
]
] | TITLE: SemDeDup: Data-efficient learning at web-scale through semantic
deduplication
ABSTRACT: Progress in machine learning has been driven in large part by massive
increases in data. However, large web-scale datasets such as LAION are largely
uncurated beyond searches for exact duplicates, potentially leaving much
redundancy. Here, we introduce SemDeDup, a method which leverages embeddings
from pre-trained models to identify and remove semantic duplicates: data pairs
which are semantically similar, but not exactly identical. Removing semantic
duplicates preserves performance and speeds up learning. Analyzing a subset of
LAION, we show that SemDeDup can remove 50% of the data with minimal
performance loss, effectively halving training time. Moreover, performance
increases out of distribution. Also, analyzing language models trained on C4, a
partially curated dataset, we show that SemDeDup improves over prior approaches
while providing efficiency gains. SemDeDup provides an example of how simple
ways of leveraging quality embeddings can be used to make models learn faster
with less data.
| no_new_dataset | 0.949059 |
2303.09638 | Lu Niu | Lu Niu, Jeremy Speth, Nathan Vance, Benjamin Sporrer, Adam Czajka,
Patrick Flynn | Full-Body Cardiovascular Sensing with Remote Photoplethysmography | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Remote photoplethysmography (rPPG) allows for noncontact monitoring of blood
volume changes from a camera by detecting minor fluctuations in reflected
light. Prior applications of rPPG focused on face videos. In this paper we
explored the feasibility of rPPG from non-face body regions such as the arms,
legs, and hands. We collected a new dataset titled Multi-Site Physiological
Monitoring (MSPM), which will be released with this paper. The dataset consists
of 90 frames per second video of exposed arms, legs, and face, along with 10
synchronized PPG recordings. We performed baseline heart rate estimation
experiments from non-face regions with several state-of-the-art rPPG
approaches, including chrominance-based (CHROM), plane-orthogonal-to-skin (POS)
and RemotePulseNet (RPNet). To our knowledge, this is the first evaluation of
the fidelity of rPPG signals simultaneously obtained from multiple regions of a
human body. Our experiments showed that skin pixels from arms, legs, and hands
are all potential sources of the blood volume pulse. The best-performing
approach, POS, achieved a mean absolute error peaking at 7.11 beats per minute
from non-facial body parts compared to 1.38 beats per minute from the face.
Additionally, we performed experiments on pulse transit time (PTT) from both
the contact PPG and rPPG signals. We found that remote PTT is possible with
moderately high frame rate video when distal locations on the body are visible.
These findings and the supporting dataset should facilitate new research on
non-face rPPG and monitoring blood flow dynamics over the whole body with a
camera.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 20:37:07 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Niu",
"Lu",
""
],
[
"Speth",
"Jeremy",
""
],
[
"Vance",
"Nathan",
""
],
[
"Sporrer",
"Benjamin",
""
],
[
"Czajka",
"Adam",
""
],
[
"Flynn",
"Patrick",
""
]
] | TITLE: Full-Body Cardiovascular Sensing with Remote Photoplethysmography
ABSTRACT: Remote photoplethysmography (rPPG) allows for noncontact monitoring of blood
volume changes from a camera by detecting minor fluctuations in reflected
light. Prior applications of rPPG focused on face videos. In this paper we
explored the feasibility of rPPG from non-face body regions such as the arms,
legs, and hands. We collected a new dataset titled Multi-Site Physiological
Monitoring (MSPM), which will be released with this paper. The dataset consists
of 90 frames per second video of exposed arms, legs, and face, along with 10
synchronized PPG recordings. We performed baseline heart rate estimation
experiments from non-face regions with several state-of-the-art rPPG
approaches, including chrominance-based (CHROM), plane-orthogonal-to-skin (POS)
and RemotePulseNet (RPNet). To our knowledge, this is the first evaluation of
the fidelity of rPPG signals simultaneously obtained from multiple regions of a
human body. Our experiments showed that skin pixels from arms, legs, and hands
are all potential sources of the blood volume pulse. The best-performing
approach, POS, achieved a mean absolute error peaking at 7.11 beats per minute
from non-facial body parts compared to 1.38 beats per minute from the face.
Additionally, we performed experiments on pulse transit time (PTT) from both
the contact PPG and rPPG signals. We found that remote PTT is possible with
moderately high frame rate video when distal locations on the body are visible.
These findings and the supporting dataset should facilitate new research on
non-face rPPG and monitoring blood flow dynamics over the whole body with a
camera.
| new_dataset | 0.964085 |
2303.09998 | Zi Wang | Shaoheng Fang, Zi Wang, Yiqi Zhong, Junhao Ge, Siheng Chen, Yanfeng
Wang | TBP-Former: Learning Temporal Bird's-Eye-View Pyramid for Joint
Perception and Prediction in Vision-Centric Autonomous Driving | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision-centric joint perception and prediction (PnP) has become an emerging
trend in autonomous driving research. It predicts the future states of the
traffic participants in the surrounding environment from raw RGB images.
However, it is still a critical challenge to synchronize features obtained at
multiple camera views and timestamps due to inevitable geometric distortions
and further exploit those spatial-temporal features. To address this issue, we
propose a temporal bird's-eye-view pyramid transformer (TBP-Former) for
vision-centric PnP, which includes two novel designs. First, a
pose-synchronized BEV encoder is proposed to map raw image inputs with any
camera pose at any time to a shared and synchronized BEV space for better
spatial-temporal synchronization. Second, a spatial-temporal pyramid
transformer is introduced to comprehensively extract multi-scale BEV features
and predict future BEV states with the support of spatial-temporal priors.
Extensive experiments on nuScenes dataset show that our proposed framework
overall outperforms all state-of-the-art vision-based prediction methods.
| [
{
"version": "v1",
"created": "Fri, 17 Mar 2023 14:20:28 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 13:58:12 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Fang",
"Shaoheng",
""
],
[
"Wang",
"Zi",
""
],
[
"Zhong",
"Yiqi",
""
],
[
"Ge",
"Junhao",
""
],
[
"Chen",
"Siheng",
""
],
[
"Wang",
"Yanfeng",
""
]
] | TITLE: TBP-Former: Learning Temporal Bird's-Eye-View Pyramid for Joint
Perception and Prediction in Vision-Centric Autonomous Driving
ABSTRACT: Vision-centric joint perception and prediction (PnP) has become an emerging
trend in autonomous driving research. It predicts the future states of the
traffic participants in the surrounding environment from raw RGB images.
However, it is still a critical challenge to synchronize features obtained at
multiple camera views and timestamps due to inevitable geometric distortions
and further exploit those spatial-temporal features. To address this issue, we
propose a temporal bird's-eye-view pyramid transformer (TBP-Former) for
vision-centric PnP, which includes two novel designs. First, a
pose-synchronized BEV encoder is proposed to map raw image inputs with any
camera pose at any time to a shared and synchronized BEV space for better
spatial-temporal synchronization. Second, a spatial-temporal pyramid
transformer is introduced to comprehensively extract multi-scale BEV features
and predict future BEV states with the support of spatial-temporal priors.
Extensive experiments on nuScenes dataset show that our proposed framework
overall outperforms all state-of-the-art vision-based prediction methods.
| no_new_dataset | 0.945801 |
2303.10659 | Ramakanth Kavuluru | Yuhang Jiang and Ramakanth Kavuluru | COVID-19 event extraction from Twitter via extractive question answering
with continuous prompts | Accepted to appear in MEDINFO 2023. Code:
https://github.com/bionlproc/twitter-covid-QA-extraction | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | As COVID-19 ravages the world, social media analytics could augment
traditional surveys in assessing how the pandemic evolves and capturing
consumer chatter that could help healthcare agencies in addressing it. This
typically involves mining disclosure events that mention testing positive for
the disease or discussions surrounding perceptions and beliefs in preventative
or treatment options. The 2020 shared task on COVID-19 event extraction
(conducted as part of the W-NUT workshop during the EMNLP conference)
introduced a new Twitter dataset for benchmarking event extraction from
COVID-19 tweets. In this paper, we cast the problem of event extraction as
extractive question answering using recent advances in continuous prompting in
language models. On the shared task test dataset, our approach leads to over 5%
absolute micro-averaged F1-score improvement over prior best results, across
all COVID-19 event slots. Our ablation study shows that continuous prompts have
a major impact on the eventual performance.
| [
{
"version": "v1",
"created": "Sun, 19 Mar 2023 13:47:56 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 15:27:08 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Jiang",
"Yuhang",
""
],
[
"Kavuluru",
"Ramakanth",
""
]
] | TITLE: COVID-19 event extraction from Twitter via extractive question answering
with continuous prompts
ABSTRACT: As COVID-19 ravages the world, social media analytics could augment
traditional surveys in assessing how the pandemic evolves and capturing
consumer chatter that could help healthcare agencies in addressing it. This
typically involves mining disclosure events that mention testing positive for
the disease or discussions surrounding perceptions and beliefs in preventative
or treatment options. The 2020 shared task on COVID-19 event extraction
(conducted as part of the W-NUT workshop during the EMNLP conference)
introduced a new Twitter dataset for benchmarking event extraction from
COVID-19 tweets. In this paper, we cast the problem of event extraction as
extractive question answering using recent advances in continuous prompting in
language models. On the shared task test dataset, our approach leads to over 5%
absolute micro-averaged F1-score improvement over prior best results, across
all COVID-19 event slots. Our ablation study shows that continuous prompts have
a major impact on the eventual performance.
| new_dataset | 0.9601 |
2303.11997 | Saizhe Ding | Saizhe Ding, Jinze Chen, Yang Wang, Yu Kang, Weiguo Song, Jie Cheng,
Yang Cao | E-MLB: Multilevel Benchmark for Event-Based Camera Denoising | null | IEEE Transactions on Multimedia, 2023 | 10.1109/TMM.2023.3260638 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Event cameras, such as dynamic vision sensors (DVS), are biologically
inspired vision sensors that have advanced over conventional cameras in high
dynamic range, low latency and low power consumption, showing great application
potential in many fields. Event cameras are more sensitive to junction leakage
current and photocurrent as they output differential signals, losing the
smoothing function of the integral imaging process in the RGB camera. The
logarithmic conversion further amplifies noise, especially in low-contrast
conditions. Recently, researchers proposed a series of datasets and evaluation
metrics but limitations remain: 1) the existing datasets are small in scale and
insufficient in noise diversity, which cannot reflect the authentic working
environments of event cameras; and 2) the existing denoising evaluation metrics
are mostly referenced evaluation metrics, relying on APS information or manual
annotation. To address the above issues, we construct a large-scale event
denoising dataset (multilevel benchmark for event denoising, E-MLB) for the
first time, which consists of 100 scenes, each with four noise levels, that is
12 times larger than the largest existing denoising dataset. We also propose
the first nonreference event denoising metric, the event structural ratio
(ESR), which measures the structural intensity of given events. ESR is inspired
by the contrast metric, but is independent of the number of events and
projection direction. Based on the proposed benchmark and ESR, we evaluate the
most representative denoising algorithms, including classic and SOTA, and
provide denoising baselines under various scenes and noise levels. The
corresponding results and codes are available at
https://github.com/KugaMaxx/cuke-emlb.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 16:31:53 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ding",
"Saizhe",
""
],
[
"Chen",
"Jinze",
""
],
[
"Wang",
"Yang",
""
],
[
"Kang",
"Yu",
""
],
[
"Song",
"Weiguo",
""
],
[
"Cheng",
"Jie",
""
],
[
"Cao",
"Yang",
""
]
] | TITLE: E-MLB: Multilevel Benchmark for Event-Based Camera Denoising
ABSTRACT: Event cameras, such as dynamic vision sensors (DVS), are biologically
inspired vision sensors that have advanced over conventional cameras in high
dynamic range, low latency and low power consumption, showing great application
potential in many fields. Event cameras are more sensitive to junction leakage
current and photocurrent as they output differential signals, losing the
smoothing function of the integral imaging process in the RGB camera. The
logarithmic conversion further amplifies noise, especially in low-contrast
conditions. Recently, researchers proposed a series of datasets and evaluation
metrics but limitations remain: 1) the existing datasets are small in scale and
insufficient in noise diversity, which cannot reflect the authentic working
environments of event cameras; and 2) the existing denoising evaluation metrics
are mostly referenced evaluation metrics, relying on APS information or manual
annotation. To address the above issues, we construct a large-scale event
denoising dataset (multilevel benchmark for event denoising, E-MLB) for the
first time, which consists of 100 scenes, each with four noise levels, that is
12 times larger than the largest existing denoising dataset. We also propose
the first nonreference event denoising metric, the event structural ratio
(ESR), which measures the structural intensity of given events. ESR is inspired
by the contrast metric, but is independent of the number of events and
projection direction. Based on the proposed benchmark and ESR, we evaluate the
most representative denoising algorithms, including classic and SOTA, and
provide denoising baselines under various scenes and noise levels. The
corresponding results and codes are available at
https://github.com/KugaMaxx/cuke-emlb.
| new_dataset | 0.972623 |
2303.12116 | Rahul Nellikkath | Rahul Nellikkath, Andreas Venzke, Mohammad Kazem Bakhshizadeh, Ilgiz
Murzakhanov and Spyros Chatzivasileiadis | Physics Informed Neural Networks for Phase Locked Loop Transient
Stability Assessment | null | null | null | null | eess.SY cs.AI cs.SY | http://creativecommons.org/licenses/by-nc-sa/4.0/ | A significant increase in renewable energy production is necessary to achieve
the UN's net-zero emission targets for 2050. Using power-electronic
controllers, such as Phase Locked Loops (PLLs), to keep grid-tied renewable
resources in synchronism with the grid can cause fast transient behavior during
grid faults leading to instability. However, assessing all the probable
scenarios is impractical, so determining the stability boundary or region of
attraction (ROA) is necessary. However, using EMT simulations or Reduced-order
models (ROMs) to accurately determine the ROA is computationally expensive.
Alternatively, Machine Learning (ML) models have been proposed as an efficient
method to predict stability. However, traditional ML algorithms require large
amounts of labeled data for training, which is computationally expensive. This
paper proposes a Physics-Informed Neural Network (PINN) architecture that
accurately predicts the nonlinear transient dynamics of a PLL controller under
fault with less labeled training data. The proposed PINN algorithm can be
incorporated into conventional simulations, accelerating EMT simulations or
ROMs by over 100 times. The PINN algorithm's performance is compared against a
ROM and an EMT simulation in PSCAD for the CIGRE benchmark model C4.49,
demonstrating its ability to accurately approximate trajectories and ROAs of a
PLL controller under varying grid impedance.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 18:09:20 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Nellikkath",
"Rahul",
""
],
[
"Venzke",
"Andreas",
""
],
[
"Bakhshizadeh",
"Mohammad Kazem",
""
],
[
"Murzakhanov",
"Ilgiz",
""
],
[
"Chatzivasileiadis",
"Spyros",
""
]
] | TITLE: Physics Informed Neural Networks for Phase Locked Loop Transient
Stability Assessment
ABSTRACT: A significant increase in renewable energy production is necessary to achieve
the UN's net-zero emission targets for 2050. Using power-electronic
controllers, such as Phase Locked Loops (PLLs), to keep grid-tied renewable
resources in synchronism with the grid can cause fast transient behavior during
grid faults leading to instability. However, assessing all the probable
scenarios is impractical, so determining the stability boundary or region of
attraction (ROA) is necessary. However, using EMT simulations or Reduced-order
models (ROMs) to accurately determine the ROA is computationally expensive.
Alternatively, Machine Learning (ML) models have been proposed as an efficient
method to predict stability. However, traditional ML algorithms require large
amounts of labeled data for training, which is computationally expensive. This
paper proposes a Physics-Informed Neural Network (PINN) architecture that
accurately predicts the nonlinear transient dynamics of a PLL controller under
fault with less labeled training data. The proposed PINN algorithm can be
incorporated into conventional simulations, accelerating EMT simulations or
ROMs by over 100 times. The PINN algorithm's performance is compared against a
ROM and an EMT simulation in PSCAD for the CIGRE benchmark model C4.49,
demonstrating its ability to accurately approximate trajectories and ROAs of a
PLL controller under varying grid impedance.
| no_new_dataset | 0.948298 |
2303.12134 | Matthias M\"uller | Diana Wofk, Ren\'e Ranftl, Matthias M\"uller, and Vladlen Koltun | Monocular Visual-Inertial Depth Estimation | Accepted for publication at ICRA'23 | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a visual-inertial depth estimation pipeline that integrates
monocular depth estimation and visual-inertial odometry to produce dense depth
estimates with metric scale. Our approach performs global scale and shift
alignment against sparse metric depth, followed by learning-based dense
alignment. We evaluate on the TartanAir and VOID datasets, observing up to 30%
reduction in inverse RMSE with dense scale alignment relative to performing
just global alignment alone. Our approach is especially competitive at low
density; with just 150 sparse metric depth points, our dense-to-dense depth
alignment method achieves over 50% lower iRMSE over sparse-to-dense depth
completion by KBNet, currently the state of the art on VOID. We demonstrate
successful zero-shot transfer from synthetic TartanAir to real-world VOID data
and perform generalization tests on NYUv2 and VCU-RVI. Our approach is modular
and is compatible with a variety of monocular depth estimation models. Video:
https://youtu.be/IMwiKwSpshQ Code: https://github.com/isl-org/VI-Depth
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 18:47:34 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wofk",
"Diana",
""
],
[
"Ranftl",
"René",
""
],
[
"Müller",
"Matthias",
""
],
[
"Koltun",
"Vladlen",
""
]
] | TITLE: Monocular Visual-Inertial Depth Estimation
ABSTRACT: We present a visual-inertial depth estimation pipeline that integrates
monocular depth estimation and visual-inertial odometry to produce dense depth
estimates with metric scale. Our approach performs global scale and shift
alignment against sparse metric depth, followed by learning-based dense
alignment. We evaluate on the TartanAir and VOID datasets, observing up to 30%
reduction in inverse RMSE with dense scale alignment relative to performing
just global alignment alone. Our approach is especially competitive at low
density; with just 150 sparse metric depth points, our dense-to-dense depth
alignment method achieves over 50% lower iRMSE over sparse-to-dense depth
completion by KBNet, currently the state of the art on VOID. We demonstrate
successful zero-shot transfer from synthetic TartanAir to real-world VOID data
and perform generalization tests on NYUv2 and VCU-RVI. Our approach is modular
and is compatible with a variety of monocular depth estimation models. Video:
https://youtu.be/IMwiKwSpshQ Code: https://github.com/isl-org/VI-Depth
| no_new_dataset | 0.953057 |
2303.12148 | Xinzi He | Xinzi He, Alan Wang, Mert R. Sabuncu | Neural Pre-Processing: A Learning Framework for End-to-end Brain MRI
Pre-processing | 8 | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Head MRI pre-processing involves converting raw images to an
intensity-normalized, skull-stripped brain in a standard coordinate space. In
this paper, we propose an end-to-end weakly supervised learning approach,
called Neural Pre-processing (NPP), for solving all three sub-tasks
simultaneously via a neural network, trained on a large dataset without
individual sub-task supervision. Because the overall objective is highly
under-constrained, we explicitly disentangle geometric-preserving intensity
mapping (skull-stripping and intensity normalization) and spatial
transformation (spatial normalization). Quantitative results show that our
model outperforms state-of-the-art methods which tackle only a single sub-task.
Our ablation experiments demonstrate the importance of the architecture design
we chose for NPP. Furthermore, NPP affords the user the flexibility to control
each of these tasks at inference time. The code and model are freely-available
at \url{https://github.com/Novestars/Neural-Pre-processing}.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 19:10:21 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"He",
"Xinzi",
""
],
[
"Wang",
"Alan",
""
],
[
"Sabuncu",
"Mert R.",
""
]
] | TITLE: Neural Pre-Processing: A Learning Framework for End-to-end Brain MRI
Pre-processing
ABSTRACT: Head MRI pre-processing involves converting raw images to an
intensity-normalized, skull-stripped brain in a standard coordinate space. In
this paper, we propose an end-to-end weakly supervised learning approach,
called Neural Pre-processing (NPP), for solving all three sub-tasks
simultaneously via a neural network, trained on a large dataset without
individual sub-task supervision. Because the overall objective is highly
under-constrained, we explicitly disentangle geometric-preserving intensity
mapping (skull-stripping and intensity normalization) and spatial
transformation (spatial normalization). Quantitative results show that our
model outperforms state-of-the-art methods which tackle only a single sub-task.
Our ablation experiments demonstrate the importance of the architecture design
we chose for NPP. Furthermore, NPP affords the user the flexibility to control
each of these tasks at inference time. The code and model are freely-available
at \url{https://github.com/Novestars/Neural-Pre-processing}.
| no_new_dataset | 0.947575 |
2303.12187 | Xiaoming Ren | Xiaoming Ren, Chao Li, Shenjian Wang, Biao Li | Practice of the conformer enhanced AUDIO-VISUAL HUBERT on Mandarin and
English | null | null | null | null | eess.AS cs.AI cs.SD | http://creativecommons.org/licenses/by/4.0/ | Considering the bimodal nature of human speech perception, lips, and teeth
movement has a pivotal role in automatic speech recognition. Benefiting from
the correlated and noise-invariant visual information, audio-visual recognition
systems enhance robustness in multiple scenarios. In previous work,
audio-visual HuBERT appears to be the finest practice incorporating modality
knowledge. This paper outlines a mixed methodology, named conformer enhanced
AV-HuBERT, boosting the AV-HuBERT system's performance a step further. Compared
with baseline AV-HuBERT, our method in the one-phase evaluation of clean and
noisy conditions achieves 7% and 16% relative WER reduction on the English AVSR
benchmark dataset LRS3. Furthermore, we establish a novel 1000h Mandarin AVSR
dataset CSTS. On top of the baseline AV-HuBERT, we exceed the WeNet ASR system
by 14% and 18% relatively on MISP and CMLR by pre-training with this dataset.
The conformer-enhanced AV-HuBERT we proposed brings 7% on MISP and 6% CER
reduction on CMLR, compared with the baseline AV-HuBERT system.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 02:10:13 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ren",
"Xiaoming",
""
],
[
"Li",
"Chao",
""
],
[
"Wang",
"Shenjian",
""
],
[
"Li",
"Biao",
""
]
] | TITLE: Practice of the conformer enhanced AUDIO-VISUAL HUBERT on Mandarin and
English
ABSTRACT: Considering the bimodal nature of human speech perception, lips, and teeth
movement has a pivotal role in automatic speech recognition. Benefiting from
the correlated and noise-invariant visual information, audio-visual recognition
systems enhance robustness in multiple scenarios. In previous work,
audio-visual HuBERT appears to be the finest practice incorporating modality
knowledge. This paper outlines a mixed methodology, named conformer enhanced
AV-HuBERT, boosting the AV-HuBERT system's performance a step further. Compared
with baseline AV-HuBERT, our method in the one-phase evaluation of clean and
noisy conditions achieves 7% and 16% relative WER reduction on the English AVSR
benchmark dataset LRS3. Furthermore, we establish a novel 1000h Mandarin AVSR
dataset CSTS. On top of the baseline AV-HuBERT, we exceed the WeNet ASR system
by 14% and 18% relatively on MISP and CMLR by pre-training with this dataset.
The conformer-enhanced AV-HuBERT we proposed brings 7% on MISP and 6% CER
reduction on CMLR, compared with the baseline AV-HuBERT system.
| new_dataset | 0.97265 |
2303.12208 | Sungwoong Kim | Sungwoong Kim, Daejin Jo, Donghoon Lee, Jongmin Kim | MAGVLT: Masked Generative Vision-and-Language Transformer | CVPR 2023 | null | null | null | cs.CV cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | While generative modeling on multimodal image-text data has been actively
developed with large-scale paired datasets, there have been limited attempts to
generate both image and text data by a single model rather than a generation of
one fixed modality conditioned on the other modality. In this paper, we explore
a unified generative vision-and-language (VL) model that can produce both
images and text sequences. Especially, we propose a generative VL transformer
based on the non-autoregressive mask prediction, named MAGVLT, and compare it
with an autoregressive generative VL transformer (ARGVLT). In comparison to
ARGVLT, the proposed MAGVLT enables bidirectional context encoding, fast
decoding by parallel token predictions in an iterative refinement, and extended
editing capabilities such as image and text infilling. For rigorous training of
our MAGVLT with image-text pairs from scratch, we combine the image-to-text,
text-to-image, and joint image-and-text mask prediction tasks. Moreover, we
devise two additional tasks based on the step-unrolled mask prediction and the
selective prediction on the mixture of two image-text pairs. Experimental
results on various downstream generation tasks of VL benchmarks show that our
MAGVLT outperforms ARGVLT by a large margin even with significant inference
speedup. Particularly, MAGVLT achieves competitive results on both zero-shot
image-to-text and text-to-image generation tasks from MS-COCO by one
moderate-sized model (fewer than 500M parameters) even without the use of
monomodal data and networks.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 21:49:39 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Kim",
"Sungwoong",
""
],
[
"Jo",
"Daejin",
""
],
[
"Lee",
"Donghoon",
""
],
[
"Kim",
"Jongmin",
""
]
] | TITLE: MAGVLT: Masked Generative Vision-and-Language Transformer
ABSTRACT: While generative modeling on multimodal image-text data has been actively
developed with large-scale paired datasets, there have been limited attempts to
generate both image and text data by a single model rather than a generation of
one fixed modality conditioned on the other modality. In this paper, we explore
a unified generative vision-and-language (VL) model that can produce both
images and text sequences. Especially, we propose a generative VL transformer
based on the non-autoregressive mask prediction, named MAGVLT, and compare it
with an autoregressive generative VL transformer (ARGVLT). In comparison to
ARGVLT, the proposed MAGVLT enables bidirectional context encoding, fast
decoding by parallel token predictions in an iterative refinement, and extended
editing capabilities such as image and text infilling. For rigorous training of
our MAGVLT with image-text pairs from scratch, we combine the image-to-text,
text-to-image, and joint image-and-text mask prediction tasks. Moreover, we
devise two additional tasks based on the step-unrolled mask prediction and the
selective prediction on the mixture of two image-text pairs. Experimental
results on various downstream generation tasks of VL benchmarks show that our
MAGVLT outperforms ARGVLT by a large margin even with significant inference
speedup. Particularly, MAGVLT achieves competitive results on both zero-shot
image-to-text and text-to-image generation tasks from MS-COCO by one
moderate-sized model (fewer than 500M parameters) even without the use of
monomodal data and networks.
| no_new_dataset | 0.950041 |
2303.12234 | Ahmad AlMughrabi | Ahmad AlMughrabi, Umair Haroon, Ricardo Marques, Petia Radeva | Pre-NeRF 360: Enriching Unbounded Appearances for Neural Radiance Fields | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Neural radiance fields (NeRF) appeared recently as a powerful tool to
generate realistic views of objects and confined areas. Still, they face
serious challenges with open scenes, where the camera has unrestricted movement
and content can appear at any distance. In such scenarios, current
NeRF-inspired models frequently yield hazy or pixelated outputs, suffer slow
training times, and might display irregularities, because of the challenging
task of reconstructing an extensive scene from a limited number of images. We
propose a new framework to boost the performance of NeRF-based architectures
yielding significantly superior outcomes compared to the prior work. Our
solution overcomes several obstacles that plagued earlier versions of NeRF,
including handling multiple video inputs, selecting keyframes, and extracting
poses from real-world frames that are ambiguous and symmetrical. Furthermore,
we applied our framework, dubbed as "Pre-NeRF 360", to enable the use of the
Nutrition5k dataset in NeRF and introduce an updated version of this dataset,
known as the N5k360 dataset.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 23:29:38 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"AlMughrabi",
"Ahmad",
""
],
[
"Haroon",
"Umair",
""
],
[
"Marques",
"Ricardo",
""
],
[
"Radeva",
"Petia",
""
]
] | TITLE: Pre-NeRF 360: Enriching Unbounded Appearances for Neural Radiance Fields
ABSTRACT: Neural radiance fields (NeRF) appeared recently as a powerful tool to
generate realistic views of objects and confined areas. Still, they face
serious challenges with open scenes, where the camera has unrestricted movement
and content can appear at any distance. In such scenarios, current
NeRF-inspired models frequently yield hazy or pixelated outputs, suffer slow
training times, and might display irregularities, because of the challenging
task of reconstructing an extensive scene from a limited number of images. We
propose a new framework to boost the performance of NeRF-based architectures
yielding significantly superior outcomes compared to the prior work. Our
solution overcomes several obstacles that plagued earlier versions of NeRF,
including handling multiple video inputs, selecting keyframes, and extracting
poses from real-world frames that are ambiguous and symmetrical. Furthermore,
we applied our framework, dubbed as "Pre-NeRF 360", to enable the use of the
Nutrition5k dataset in NeRF and introduce an updated version of this dataset,
known as the N5k360 dataset.
| new_dataset | 0.95469 |
2303.12238 | Yanshen Sun | Yanshen Sun, Kaiqun Fu, and Chang-Tien Lu | DG-Trans: Dual-level Graph Transformer for Spatiotemporal Incident
Impact Prediction on Traffic Networks | null | null | null | null | cs.LG cs.SI | http://creativecommons.org/licenses/by/4.0/ | The prompt estimation of traffic incident impacts can guide commuters in
their trip planning and improve the resilience of transportation agencies'
decision-making on resilience. However, it is more challenging than node-level
and graph-level forecasting tasks, as it requires extracting the anomaly
subgraph or sub-time-series from dynamic graphs. In this paper, we propose
DG-Trans, a novel traffic incident impact prediction framework, to foresee the
impact of traffic incidents through dynamic graph learning. The proposed
framework contains a dual-level spatial transformer and an
importance-score-based temporal transformer, and the performance of this
framework is justified by two newly constructed benchmark datasets. The
dual-level spatial transformer removes unnecessary edges between nodes to
isolate the affected subgraph from the other nodes. Meanwhile, the
importance-score-based temporal transformer identifies abnormal changes in node
features, causing the predictions to rely more on measurement changes after the
incident occurs. Therefore, DG-Trans is equipped with dual abilities that
extract spatiotemporal dependency and identify anomaly nodes affected by
incidents while removing noise introduced by benign nodes. Extensive
experiments on real-world datasets verify that DG-Trans outperforms the
existing state-of-the-art methods, especially in extracting spatiotemporal
dependency patterns and predicting traffic accident impacts. It offers
promising potential for traffic incident management systems.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 23:44:09 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Sun",
"Yanshen",
""
],
[
"Fu",
"Kaiqun",
""
],
[
"Lu",
"Chang-Tien",
""
]
] | TITLE: DG-Trans: Dual-level Graph Transformer for Spatiotemporal Incident
Impact Prediction on Traffic Networks
ABSTRACT: The prompt estimation of traffic incident impacts can guide commuters in
their trip planning and improve the resilience of transportation agencies'
decision-making on resilience. However, it is more challenging than node-level
and graph-level forecasting tasks, as it requires extracting the anomaly
subgraph or sub-time-series from dynamic graphs. In this paper, we propose
DG-Trans, a novel traffic incident impact prediction framework, to foresee the
impact of traffic incidents through dynamic graph learning. The proposed
framework contains a dual-level spatial transformer and an
importance-score-based temporal transformer, and the performance of this
framework is justified by two newly constructed benchmark datasets. The
dual-level spatial transformer removes unnecessary edges between nodes to
isolate the affected subgraph from the other nodes. Meanwhile, the
importance-score-based temporal transformer identifies abnormal changes in node
features, causing the predictions to rely more on measurement changes after the
incident occurs. Therefore, DG-Trans is equipped with dual abilities that
extract spatiotemporal dependency and identify anomaly nodes affected by
incidents while removing noise introduced by benign nodes. Extensive
experiments on real-world datasets verify that DG-Trans outperforms the
existing state-of-the-art methods, especially in extracting spatiotemporal
dependency patterns and predicting traffic accident impacts. It offers
promising potential for traffic incident management systems.
| no_new_dataset | 0.854095 |
2303.12241 | Shiqiang Du | Kaiwu Zhang, Shiqiang Du, Baokai Liu, and Shengxia Gao | Preventing Dimensional Collapse of Incomplete Multi-View Clustering via
Direct Contrastive Learning | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Incomplete multi-view clustering (IMVC) is an unsupervised approach, among
which IMVC via contrastive learning has received attention due to its excellent
performance. The previous methods have the following problems: 1) Over-reliance
on additional projection heads when solving the dimensional collapse problem in
which latent features are only valid in lower-dimensional subspaces during
clustering. However, many parameters in the projection heads are unnecessary.
2) The recovered view contain inconsistent private information and useless
private information will mislead the learning of common semantics due to
consistent learning and reconstruction learning on the same feature. To address
the above issues, we propose a novel incomplete multi-view contrastive
clustering framework. This framework directly optimizes the latent feature
subspace, utilizes the learned feature vectors and their sub-vectors for
reconstruction learning and consistency learning, thereby effectively avoiding
dimensional collapse without relying on projection heads. Since reconstruction
loss and contrastive loss are performed on different features, the adverse
effect of useless private information is reduced. For the incomplete data, the
missing information is recovered by the cross-view prediction mechanism and the
inconsistent information from different views is discarded by the minimum
conditional entropy to further avoid the influence of private information.
Extensive experimental results of the method on 5 public datasets show that the
method achieves state-of-the-art clustering results.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 00:21:50 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhang",
"Kaiwu",
""
],
[
"Du",
"Shiqiang",
""
],
[
"Liu",
"Baokai",
""
],
[
"Gao",
"Shengxia",
""
]
] | TITLE: Preventing Dimensional Collapse of Incomplete Multi-View Clustering via
Direct Contrastive Learning
ABSTRACT: Incomplete multi-view clustering (IMVC) is an unsupervised approach, among
which IMVC via contrastive learning has received attention due to its excellent
performance. The previous methods have the following problems: 1) Over-reliance
on additional projection heads when solving the dimensional collapse problem in
which latent features are only valid in lower-dimensional subspaces during
clustering. However, many parameters in the projection heads are unnecessary.
2) The recovered view contain inconsistent private information and useless
private information will mislead the learning of common semantics due to
consistent learning and reconstruction learning on the same feature. To address
the above issues, we propose a novel incomplete multi-view contrastive
clustering framework. This framework directly optimizes the latent feature
subspace, utilizes the learned feature vectors and their sub-vectors for
reconstruction learning and consistency learning, thereby effectively avoiding
dimensional collapse without relying on projection heads. Since reconstruction
loss and contrastive loss are performed on different features, the adverse
effect of useless private information is reduced. For the incomplete data, the
missing information is recovered by the cross-view prediction mechanism and the
inconsistent information from different views is discarded by the minimum
conditional entropy to further avoid the influence of private information.
Extensive experimental results of the method on 5 public datasets show that the
method achieves state-of-the-art clustering results.
| no_new_dataset | 0.950915 |
2303.12246 | Heng Yang | Heng Yang, Marco Pavone | Object Pose Estimation with Statistical Guarantees: Conformal Keypoint
Detection and Geometric Uncertainty Propagation | Accepted at CVPR 2023 as a highlight paper | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by/4.0/ | The two-stage object pose estimation paradigm first detects semantic
keypoints on the image and then estimates the 6D pose by minimizing
reprojection errors. Despite performing well on standard benchmarks, existing
techniques offer no provable guarantees on the quality and uncertainty of the
estimation. In this paper, we inject two fundamental changes, namely conformal
keypoint detection and geometric uncertainty propagation, into the two-stage
paradigm and propose the first pose estimator that endows an estimation with
provable and computable worst-case error bounds. On one hand, conformal
keypoint detection applies the statistical machinery of inductive conformal
prediction to convert heuristic keypoint detections into circular or elliptical
prediction sets that cover the groundtruth keypoints with a user-specified
marginal probability (e.g., 90%). Geometric uncertainty propagation, on the
other, propagates the geometric constraints on the keypoints to the 6D object
pose, leading to a Pose UnceRtainty SEt (PURSE) that guarantees coverage of the
groundtruth pose with the same probability. The PURSE, however, is a nonconvex
set that does not directly lead to estimated poses and uncertainties.
Therefore, we develop RANdom SAmple averaGing (RANSAG) to compute an average
pose and apply semidefinite relaxation to upper bound the worst-case errors
between the average pose and the groundtruth. On the LineMOD Occlusion dataset
we demonstrate: (i) the PURSE covers the groundtruth with valid probabilities;
(ii) the worst-case error bounds provide correct uncertainty quantification;
and (iii) the average pose achieves better or similar accuracy as
representative methods based on sparse keypoints.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 00:55:53 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Yang",
"Heng",
""
],
[
"Pavone",
"Marco",
""
]
] | TITLE: Object Pose Estimation with Statistical Guarantees: Conformal Keypoint
Detection and Geometric Uncertainty Propagation
ABSTRACT: The two-stage object pose estimation paradigm first detects semantic
keypoints on the image and then estimates the 6D pose by minimizing
reprojection errors. Despite performing well on standard benchmarks, existing
techniques offer no provable guarantees on the quality and uncertainty of the
estimation. In this paper, we inject two fundamental changes, namely conformal
keypoint detection and geometric uncertainty propagation, into the two-stage
paradigm and propose the first pose estimator that endows an estimation with
provable and computable worst-case error bounds. On one hand, conformal
keypoint detection applies the statistical machinery of inductive conformal
prediction to convert heuristic keypoint detections into circular or elliptical
prediction sets that cover the groundtruth keypoints with a user-specified
marginal probability (e.g., 90%). Geometric uncertainty propagation, on the
other, propagates the geometric constraints on the keypoints to the 6D object
pose, leading to a Pose UnceRtainty SEt (PURSE) that guarantees coverage of the
groundtruth pose with the same probability. The PURSE, however, is a nonconvex
set that does not directly lead to estimated poses and uncertainties.
Therefore, we develop RANdom SAmple averaGing (RANSAG) to compute an average
pose and apply semidefinite relaxation to upper bound the worst-case errors
between the average pose and the groundtruth. On the LineMOD Occlusion dataset
we demonstrate: (i) the PURSE covers the groundtruth with valid probabilities;
(ii) the worst-case error bounds provide correct uncertainty quantification;
and (iii) the average pose achieves better or similar accuracy as
representative methods based on sparse keypoints.
| no_new_dataset | 0.947235 |
2303.12254 | Nathan Braswell | Nathan Braswell, Sharjeel Khan, Santosh Pande | Practical compilation of fexprs using partial evaluation: Fexprs can
performantly replace macros in purely-functional Lisp | 26 pages, 23 figures (incl appendix), submitted to ICFP23 | null | null | null | cs.PL | http://creativecommons.org/licenses/by/4.0/ | Macros are a common part of Lisp languages, and one of their most lauded
features. Much research has gone into making macros both safer and more
powerful resulting in developments in multiple areas, including maintaining
hygiene, and typed program staging. However, macros do suffer from various
downsides, including being second-class. Particularly egregious for eager
functional programming, they are unable to be passed to higher-order functions
or freely composed. Fexprs, as reformulated by John Shutt, provide a
first-class and more powerful alternative to macros that meshes well with pure
functional programming. Unfortunately, naive execution of fexprs is much slower
than macros due to re-executing unoptimized operative combiner code at runtime
that, in a macro-based language, would have been expanded and then optimized at
compile time. To show that fexprs can be practical replacements for macros, we
formulate a small purely functional fexpr based Lisp, Kraken, with an online
partial evaluation and compilation framework that supports first-class,
partially-static-data environments and can completely optimize away fexprs that
are used and written in the style of macros. We show our partial evaluation and
compilation framework produces code that is more than 70,000 times faster than
naive interpretation due to the elimination of repeated work and exposure of
static information enabling additional optimization. In addition, our Kraken
compiler performs better compared to existing interpreted languages that
support fexprs, including improving on NewLisp's fexpr performance by 233x on
one benchmark.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 01:45:03 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Braswell",
"Nathan",
""
],
[
"Khan",
"Sharjeel",
""
],
[
"Pande",
"Santosh",
""
]
] | TITLE: Practical compilation of fexprs using partial evaluation: Fexprs can
performantly replace macros in purely-functional Lisp
ABSTRACT: Macros are a common part of Lisp languages, and one of their most lauded
features. Much research has gone into making macros both safer and more
powerful resulting in developments in multiple areas, including maintaining
hygiene, and typed program staging. However, macros do suffer from various
downsides, including being second-class. Particularly egregious for eager
functional programming, they are unable to be passed to higher-order functions
or freely composed. Fexprs, as reformulated by John Shutt, provide a
first-class and more powerful alternative to macros that meshes well with pure
functional programming. Unfortunately, naive execution of fexprs is much slower
than macros due to re-executing unoptimized operative combiner code at runtime
that, in a macro-based language, would have been expanded and then optimized at
compile time. To show that fexprs can be practical replacements for macros, we
formulate a small purely functional fexpr based Lisp, Kraken, with an online
partial evaluation and compilation framework that supports first-class,
partially-static-data environments and can completely optimize away fexprs that
are used and written in the style of macros. We show our partial evaluation and
compilation framework produces code that is more than 70,000 times faster than
naive interpretation due to the elimination of repeated work and exposure of
static information enabling additional optimization. In addition, our Kraken
compiler performs better compared to existing interpreted languages that
support fexprs, including improving on NewLisp's fexpr performance by 233x on
one benchmark.
| no_new_dataset | 0.933431 |
2303.12267 | Puning Yang | Puning Yang, Jian Liang, Jie Cao, Ran He | AUTO: Adaptive Outlier Optimization for Online Test-Time OOD Detection | 15 pages | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Out-of-distribution (OOD) detection is a crucial aspect of deploying machine
learning models in open-world applications. Empirical evidence suggests that
training with auxiliary outliers substantially improves OOD detection. However,
such outliers typically exhibit a distribution gap compared to the test OOD
data and do not cover all possible test OOD scenarios. Additionally,
incorporating these outliers introduces additional training burdens. In this
paper, we introduce a novel paradigm called test-time OOD detection, which
utilizes unlabeled online data directly at test time to improve OOD detection
performance. While this paradigm is efficient, it also presents challenges such
as catastrophic forgetting. To address these challenges, we propose adaptive
outlier optimization (AUTO), which consists of an in-out-aware filter, an ID
memory bank, and a semantically-consistent objective. AUTO adaptively mines
pseudo-ID and pseudo-OOD samples from test data, utilizing them to optimize
networks in real time during inference. Extensive results on CIFAR-10,
CIFAR-100, and ImageNet benchmarks demonstrate that AUTO significantly enhances
OOD detection performance.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 02:28:54 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Yang",
"Puning",
""
],
[
"Liang",
"Jian",
""
],
[
"Cao",
"Jie",
""
],
[
"He",
"Ran",
""
]
] | TITLE: AUTO: Adaptive Outlier Optimization for Online Test-Time OOD Detection
ABSTRACT: Out-of-distribution (OOD) detection is a crucial aspect of deploying machine
learning models in open-world applications. Empirical evidence suggests that
training with auxiliary outliers substantially improves OOD detection. However,
such outliers typically exhibit a distribution gap compared to the test OOD
data and do not cover all possible test OOD scenarios. Additionally,
incorporating these outliers introduces additional training burdens. In this
paper, we introduce a novel paradigm called test-time OOD detection, which
utilizes unlabeled online data directly at test time to improve OOD detection
performance. While this paradigm is efficient, it also presents challenges such
as catastrophic forgetting. To address these challenges, we propose adaptive
outlier optimization (AUTO), which consists of an in-out-aware filter, an ID
memory bank, and a semantically-consistent objective. AUTO adaptively mines
pseudo-ID and pseudo-OOD samples from test data, utilizing them to optimize
networks in real time during inference. Extensive results on CIFAR-10,
CIFAR-100, and ImageNet benchmarks demonstrate that AUTO significantly enhances
OOD detection performance.
| no_new_dataset | 0.947088 |
2303.12269 | Behnam Ghavami | Eduardo Rhod, Behnam Ghavami, Zhenman Fang, Lesley Shannon | A Cycle-Accurate Soft Error Vulnerability Analysis Framework for
FPGA-based Designs | null | null | null | null | cs.AR | http://creativecommons.org/licenses/by/4.0/ | Many aerospace and automotive applications use FPGAs in their designs due to
their low power and reconfigurability requirements. Meanwhile, such
applications also pose a high standard on system reliability, which makes the
early-stage reliability analysis for FPGA-based designs very critical.
In this paper, we present a framework that enables fast and accurate
early-stage analysis of soft error vulnerability for small FPGA-based designs.
Our framework first extracts the post-synthesis netlist from an FPGA design.
Then it inserts the bit-flip configuration faults into the design netlist using
our proposed interface software. After that, it seamlessly feeds the golden
copy and fault copies of the netlist into the open source simulator Verilator
for cycle-accurate simulation. Finally, it generates a histogram of
vulnerability scores of the original design to guide the reliability analysis.
Experimental results show that our framework runs up to 53x faster than the
Xilinx Vivado fault simulation with cycle-level accuracy, when analyzing the
injected bit-flip faults on the ITC'99 benchmarks.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 02:35:07 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Rhod",
"Eduardo",
""
],
[
"Ghavami",
"Behnam",
""
],
[
"Fang",
"Zhenman",
""
],
[
"Shannon",
"Lesley",
""
]
] | TITLE: A Cycle-Accurate Soft Error Vulnerability Analysis Framework for
FPGA-based Designs
ABSTRACT: Many aerospace and automotive applications use FPGAs in their designs due to
their low power and reconfigurability requirements. Meanwhile, such
applications also pose a high standard on system reliability, which makes the
early-stage reliability analysis for FPGA-based designs very critical.
In this paper, we present a framework that enables fast and accurate
early-stage analysis of soft error vulnerability for small FPGA-based designs.
Our framework first extracts the post-synthesis netlist from an FPGA design.
Then it inserts the bit-flip configuration faults into the design netlist using
our proposed interface software. After that, it seamlessly feeds the golden
copy and fault copies of the netlist into the open source simulator Verilator
for cycle-accurate simulation. Finally, it generates a histogram of
vulnerability scores of the original design to guide the reliability analysis.
Experimental results show that our framework runs up to 53x faster than the
Xilinx Vivado fault simulation with cycle-level accuracy, when analyzing the
injected bit-flip faults on the ITC'99 benchmarks.
| no_new_dataset | 0.946794 |
2303.12279 | Eason Chen | Eason Chen | Generate labeled training data using Prompt Programming and GPT-3. An
example of Big Five Personality Classification | 6 pages, pre-print | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | We generated 25000 conversations labeled with Big Five Personality traits
using prompt programming at GPT-3. Then we train Big Five classification models
with these data and evaluate them with 2500 data from generated dialogues and
real conversational datasets labeled in Big Five by human annotators. The
results indicated that this approach is promising for creating effective
training data. We then compare the performance by different training approaches
and models. Our results suggest that using Adapter-Transformers and transfer
learning from pre-trained RoBERTa sentiment analysis model will perform best
with the generated data. Our best model obtained an accuracy of 0.71 in
generated data and 0.65 in real datasets. Finally, we discuss this approach's
potential limitations and confidence metric.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 03:12:40 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Chen",
"Eason",
""
]
] | TITLE: Generate labeled training data using Prompt Programming and GPT-3. An
example of Big Five Personality Classification
ABSTRACT: We generated 25000 conversations labeled with Big Five Personality traits
using prompt programming at GPT-3. Then we train Big Five classification models
with these data and evaluate them with 2500 data from generated dialogues and
real conversational datasets labeled in Big Five by human annotators. The
results indicated that this approach is promising for creating effective
training data. We then compare the performance by different training approaches
and models. Our results suggest that using Adapter-Transformers and transfer
learning from pre-trained RoBERTa sentiment analysis model will perform best
with the generated data. Our best model obtained an accuracy of 0.71 in
generated data and 0.65 in real datasets. Finally, we discuss this approach's
potential limitations and confidence metric.
| no_new_dataset | 0.952042 |
2303.12281 | Nicholas Kuo | Nicholas I-Hsien Kuo, Louisa Jorm, Sebastiano Barbieri | Synthetic Health-related Longitudinal Data with Mixed-type Variables
Generated using Diffusion Models | null | null | null | null | cs.LG q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | This paper presents a novel approach to simulating electronic health records
(EHRs) using diffusion probabilistic models (DPMs). Specifically, we
demonstrate the effectiveness of DPMs in synthesising longitudinal EHRs that
capture mixed-type variables, including numeric, binary, and categorical
variables. To our knowledge, this represents the first use of DPMs for this
purpose. We compared our DPM-simulated datasets to previous state-of-the-art
results based on generative adversarial networks (GANs) for two clinical
applications: acute hypotension and human immunodeficiency virus (ART for HIV).
Given the lack of similar previous studies in DPMs, a core component of our
work involves exploring the advantages and caveats of employing DPMs across a
wide range of aspects. In addition to assessing the realism of the synthetic
datasets, we also trained reinforcement learning (RL) agents on the synthetic
data to evaluate their utility for supporting the development of downstream
machine learning models. Finally, we estimated that our DPM-simulated datasets
are secure and posed a low patient exposure risk for public access.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 03:15:33 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Kuo",
"Nicholas I-Hsien",
""
],
[
"Jorm",
"Louisa",
""
],
[
"Barbieri",
"Sebastiano",
""
]
] | TITLE: Synthetic Health-related Longitudinal Data with Mixed-type Variables
Generated using Diffusion Models
ABSTRACT: This paper presents a novel approach to simulating electronic health records
(EHRs) using diffusion probabilistic models (DPMs). Specifically, we
demonstrate the effectiveness of DPMs in synthesising longitudinal EHRs that
capture mixed-type variables, including numeric, binary, and categorical
variables. To our knowledge, this represents the first use of DPMs for this
purpose. We compared our DPM-simulated datasets to previous state-of-the-art
results based on generative adversarial networks (GANs) for two clinical
applications: acute hypotension and human immunodeficiency virus (ART for HIV).
Given the lack of similar previous studies in DPMs, a core component of our
work involves exploring the advantages and caveats of employing DPMs across a
wide range of aspects. In addition to assessing the realism of the synthetic
datasets, we also trained reinforcement learning (RL) agents on the synthetic
data to evaluate their utility for supporting the development of downstream
machine learning models. Finally, we estimated that our DPM-simulated datasets
are secure and posed a low patient exposure risk for public access.
| no_new_dataset | 0.949012 |
2303.12289 | Qiming Ye Mr | Qiming Ye, Yuxiang Feng, Jose Javier Escribano Macias, Marc Stettler,
Panagiotis Angeloudis | Adaptive Road Configurations for Improved Autonomous Vehicle-Pedestrian
Interactions using Reinforcement Learning | 11 pages, 7 figures, Copyright \c{opyright} 2023, IEEE | IEEE Transactions on Intelligent Transportation Systems, vol. 24,
no. 2, pp. 2024-2034, Feb. 2023 | 10.1109/TITS.2022.3220110 | null | cs.LG cs.AI cs.RO cs.SY eess.SY math.DS math.OC | http://creativecommons.org/licenses/by/4.0/ | The deployment of Autonomous Vehicles (AVs) poses considerable challenges and
unique opportunities for the design and management of future urban road
infrastructure. In light of this disruptive transformation, the Right-Of-Way
(ROW) composition of road space has the potential to be renewed. Design
approaches and intelligent control models have been proposed to address this
problem, but we lack an operational framework that can dynamically generate ROW
plans for AVs and pedestrians in response to real-time demand. Based on
microscopic traffic simulation, this study explores Reinforcement Learning (RL)
methods for evolving ROW compositions. We implement a centralised paradigm and
a distributive learning paradigm to separately perform the dynamic control on
several road network configurations. Experimental results indicate that the
algorithms have the potential to improve traffic flow efficiency and allocate
more space for pedestrians. Furthermore, the distributive learning algorithm
outperforms its centralised counterpart regarding computational cost (49.55\%),
benchmark rewards (25.35\%), best cumulative rewards (24.58\%), optimal actions
(13.49\%) and rate of convergence. This novel road management technique could
potentially contribute to the flow-adaptive and active mobility-friendly
streets in the AVs era.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 03:42:39 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ye",
"Qiming",
""
],
[
"Feng",
"Yuxiang",
""
],
[
"Macias",
"Jose Javier Escribano",
""
],
[
"Stettler",
"Marc",
""
],
[
"Angeloudis",
"Panagiotis",
""
]
] | TITLE: Adaptive Road Configurations for Improved Autonomous Vehicle-Pedestrian
Interactions using Reinforcement Learning
ABSTRACT: The deployment of Autonomous Vehicles (AVs) poses considerable challenges and
unique opportunities for the design and management of future urban road
infrastructure. In light of this disruptive transformation, the Right-Of-Way
(ROW) composition of road space has the potential to be renewed. Design
approaches and intelligent control models have been proposed to address this
problem, but we lack an operational framework that can dynamically generate ROW
plans for AVs and pedestrians in response to real-time demand. Based on
microscopic traffic simulation, this study explores Reinforcement Learning (RL)
methods for evolving ROW compositions. We implement a centralised paradigm and
a distributive learning paradigm to separately perform the dynamic control on
several road network configurations. Experimental results indicate that the
algorithms have the potential to improve traffic flow efficiency and allocate
more space for pedestrians. Furthermore, the distributive learning algorithm
outperforms its centralised counterpart regarding computational cost (49.55\%),
benchmark rewards (25.35\%), best cumulative rewards (24.58\%), optimal actions
(13.49\%) and rate of convergence. This novel road management technique could
potentially contribute to the flow-adaptive and active mobility-friendly
streets in the AVs era.
| no_new_dataset | 0.945399 |
2303.12291 | Jiaheng Wei | Jiaheng Wei, Zhaowei Zhu, Gang Niu, Tongliang Liu, Sijia Liu, Masashi
Sugiyama, and Yang Liu | Fairness Improves Learning from Noisily Labeled Long-Tailed Data | Paper under review | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Both long-tailed and noisily labeled data frequently appear in real-world
applications and impose significant challenges for learning. Most prior works
treat either problem in an isolated way and do not explicitly consider the
coupling effects of the two. Our empirical observation reveals that such
solutions fail to consistently improve the learning when the dataset is
long-tailed with label noise. Moreover, with the presence of label noise,
existing methods do not observe universal improvements across different
sub-populations; in other words, some sub-populations enjoyed the benefits of
improved accuracy at the cost of hurting others. Based on these observations,
we introduce the Fairness Regularizer (FR), inspired by regularizing the
performance gap between any two sub-populations. We show that the introduced
fairness regularizer improves the performances of sub-populations on the tail
and the overall learning performance. Extensive experiments demonstrate the
effectiveness of the proposed solution when complemented with certain existing
popular robust or class-balanced methods.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 03:46:51 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wei",
"Jiaheng",
""
],
[
"Zhu",
"Zhaowei",
""
],
[
"Niu",
"Gang",
""
],
[
"Liu",
"Tongliang",
""
],
[
"Liu",
"Sijia",
""
],
[
"Sugiyama",
"Masashi",
""
],
[
"Liu",
"Yang",
""
]
] | TITLE: Fairness Improves Learning from Noisily Labeled Long-Tailed Data
ABSTRACT: Both long-tailed and noisily labeled data frequently appear in real-world
applications and impose significant challenges for learning. Most prior works
treat either problem in an isolated way and do not explicitly consider the
coupling effects of the two. Our empirical observation reveals that such
solutions fail to consistently improve the learning when the dataset is
long-tailed with label noise. Moreover, with the presence of label noise,
existing methods do not observe universal improvements across different
sub-populations; in other words, some sub-populations enjoyed the benefits of
improved accuracy at the cost of hurting others. Based on these observations,
we introduce the Fairness Regularizer (FR), inspired by regularizing the
performance gap between any two sub-populations. We show that the introduced
fairness regularizer improves the performances of sub-populations on the tail
and the overall learning performance. Extensive experiments demonstrate the
effectiveness of the proposed solution when complemented with certain existing
popular robust or class-balanced methods.
| no_new_dataset | 0.947817 |
2303.12296 | Yu Qiao | Yu Qiao, Seong-Bae Park, Sun Moo Kang, and Choong Seon Hong | Prototype Helps Federated Learning: Towards Faster Convergence | 3 pages, 3 figures | null | null | null | cs.LG cs.AI cs.MM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Federated learning (FL) is a distributed machine learning technique in which
multiple clients cooperate to train a shared model without exchanging their raw
data. However, heterogeneity of data distribution among clients usually leads
to poor model inference. In this paper, a prototype-based federated learning
framework is proposed, which can achieve better inference performance with only
a few changes to the last global iteration of the typical federated learning
process. In the last iteration, the server aggregates the prototypes
transmitted from distributed clients and then sends them back to local clients
for their respective model inferences. Experiments on two baseline datasets
show that our proposal can achieve higher accuracy (at least 1%) and relatively
efficient communication than two popular baselines under different
heterogeneous settings.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 04:06:29 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Qiao",
"Yu",
""
],
[
"Park",
"Seong-Bae",
""
],
[
"Kang",
"Sun Moo",
""
],
[
"Hong",
"Choong Seon",
""
]
] | TITLE: Prototype Helps Federated Learning: Towards Faster Convergence
ABSTRACT: Federated learning (FL) is a distributed machine learning technique in which
multiple clients cooperate to train a shared model without exchanging their raw
data. However, heterogeneity of data distribution among clients usually leads
to poor model inference. In this paper, a prototype-based federated learning
framework is proposed, which can achieve better inference performance with only
a few changes to the last global iteration of the typical federated learning
process. In the last iteration, the server aggregates the prototypes
transmitted from distributed clients and then sends them back to local clients
for their respective model inferences. Experiments on two baseline datasets
show that our proposal can achieve higher accuracy (at least 1%) and relatively
efficient communication than two popular baselines under different
heterogeneous settings.
| no_new_dataset | 0.950227 |
2303.12304 | Jiahao Bao | Jiahao Bao, Kaiqiang Chen, Xian Sun, Liangjin Zhao, Wenhui Diao,
Menglong Yan | SiamTHN: Siamese Target Highlight Network for Visual Tracking | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Siamese network based trackers develop rapidly in the field of visual object
tracking in recent years. The majority of siamese network based trackers now in
use treat each channel in the feature maps generated by the backbone network
equally, making the similarity response map sensitive to background influence
and hence challenging to focus on the target region. Additionally, there are no
structural links between the classification and regression branches in these
trackers, and the two branches are optimized separately during training.
Therefore, there is a misalignment between the classification and regression
branches, which results in less accurate tracking results. In this paper, a
Target Highlight Module is proposed to help the generated similarity response
maps to be more focused on the target region. To reduce the misalignment and
produce more precise tracking results, we propose a corrective loss to train
the model. The two branches of the model are jointly tuned with the use of
corrective loss to produce more reliable prediction results. Experiments on 5
challenging benchmark datasets reveal that the method outperforms current
models in terms of performance, and runs at 38 fps, proving its effectiveness
and efficiency.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 04:33:02 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Bao",
"Jiahao",
""
],
[
"Chen",
"Kaiqiang",
""
],
[
"Sun",
"Xian",
""
],
[
"Zhao",
"Liangjin",
""
],
[
"Diao",
"Wenhui",
""
],
[
"Yan",
"Menglong",
""
]
] | TITLE: SiamTHN: Siamese Target Highlight Network for Visual Tracking
ABSTRACT: Siamese network based trackers develop rapidly in the field of visual object
tracking in recent years. The majority of siamese network based trackers now in
use treat each channel in the feature maps generated by the backbone network
equally, making the similarity response map sensitive to background influence
and hence challenging to focus on the target region. Additionally, there are no
structural links between the classification and regression branches in these
trackers, and the two branches are optimized separately during training.
Therefore, there is a misalignment between the classification and regression
branches, which results in less accurate tracking results. In this paper, a
Target Highlight Module is proposed to help the generated similarity response
maps to be more focused on the target region. To reduce the misalignment and
produce more precise tracking results, we propose a corrective loss to train
the model. The two branches of the model are jointly tuned with the use of
corrective loss to produce more reliable prediction results. Experiments on 5
challenging benchmark datasets reveal that the method outperforms current
models in terms of performance, and runs at 38 fps, proving its effectiveness
and efficiency.
| no_new_dataset | 0.952264 |
2303.12311 | Jun Li | Jun Li, Che Liu, Sibo Cheng, Rossella Arcucci, Shenda Hong | Frozen Language Model Helps ECG Zero-Shot Learning | null | null | null | null | cs.LG cs.AI q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | The electrocardiogram (ECG) is one of the most commonly used non-invasive,
convenient medical monitoring tools that assist in the clinical diagnosis of
heart diseases. Recently, deep learning (DL) techniques, particularly
self-supervised learning (SSL), have demonstrated great potential in the
classification of ECG. SSL pre-training has achieved competitive performance
with only a small amount of annotated data after fine-tuning. However, current
SSL methods rely on the availability of annotated data and are unable to
predict labels not existing in fine-tuning datasets. To address this challenge,
we propose Multimodal ECG-Text Self-supervised pre-training (METS), the first
work to utilize the auto-generated clinical reports to guide ECG SSL
pre-training. We use a trainable ECG encoder and a frozen language model to
embed paired ECG and automatically machine-generated clinical reports
separately. The SSL aims to maximize the similarity between paired ECG and
auto-generated report while minimize the similarity between ECG and other
reports. In downstream classification tasks, METS achieves around 10%
improvement in performance without using any annotated data via zero-shot
classification, compared to other supervised and SSL baselines that rely on
annotated data. Furthermore, METS achieves the highest recall and F1 scores on
the MIT-BIH dataset, despite MIT-BIH containing different classes of ECG
compared to the pre-trained dataset. The extensive experiments have
demonstrated the advantages of using ECG-Text multimodal self-supervised
learning in terms of generalizability, effectiveness, and efficiency.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 05:01:14 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Li",
"Jun",
""
],
[
"Liu",
"Che",
""
],
[
"Cheng",
"Sibo",
""
],
[
"Arcucci",
"Rossella",
""
],
[
"Hong",
"Shenda",
""
]
] | TITLE: Frozen Language Model Helps ECG Zero-Shot Learning
ABSTRACT: The electrocardiogram (ECG) is one of the most commonly used non-invasive,
convenient medical monitoring tools that assist in the clinical diagnosis of
heart diseases. Recently, deep learning (DL) techniques, particularly
self-supervised learning (SSL), have demonstrated great potential in the
classification of ECG. SSL pre-training has achieved competitive performance
with only a small amount of annotated data after fine-tuning. However, current
SSL methods rely on the availability of annotated data and are unable to
predict labels not existing in fine-tuning datasets. To address this challenge,
we propose Multimodal ECG-Text Self-supervised pre-training (METS), the first
work to utilize the auto-generated clinical reports to guide ECG SSL
pre-training. We use a trainable ECG encoder and a frozen language model to
embed paired ECG and automatically machine-generated clinical reports
separately. The SSL aims to maximize the similarity between paired ECG and
auto-generated report while minimize the similarity between ECG and other
reports. In downstream classification tasks, METS achieves around 10%
improvement in performance without using any annotated data via zero-shot
classification, compared to other supervised and SSL baselines that rely on
annotated data. Furthermore, METS achieves the highest recall and F1 scores on
the MIT-BIH dataset, despite MIT-BIH containing different classes of ECG
compared to the pre-trained dataset. The extensive experiments have
demonstrated the advantages of using ECG-Text multimodal self-supervised
learning in terms of generalizability, effectiveness, and efficiency.
| no_new_dataset | 0.950641 |
2303.12316 | Sumanta Mukherjee | Vikas C. Raykar, Arindam Jati, Sumanta Mukherjee, Nupur Aggarwal,
Kanthi Sarpatwar, Giridhar Ganapavarapu, Roman Vaculin | TsSHAP: Robust model agnostic feature-based explainability for time
series forecasting | 11 pages, 8 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A trustworthy machine learning model should be accurate as well as
explainable. Understanding why a model makes a certain decision defines the
notion of explainability. While various flavors of explainability have been
well-studied in supervised learning paradigms like classification and
regression, literature on explainability for time series forecasting is
relatively scarce.
In this paper, we propose a feature-based explainability algorithm, TsSHAP,
that can explain the forecast of any black-box forecasting model. The method is
agnostic of the forecasting model and can provide explanations for a forecast
in terms of interpretable features defined by the user a prior.
The explanations are in terms of the SHAP values obtained by applying the
TreeSHAP algorithm on a surrogate model that learns a mapping between the
interpretable feature space and the forecast of the black-box model.
Moreover, we formalize the notion of local, semi-local, and global
explanations in the context of time series forecasting, which can be useful in
several scenarios. We validate the efficacy and robustness of TsSHAP through
extensive experiments on multiple datasets.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 05:14:36 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Raykar",
"Vikas C.",
""
],
[
"Jati",
"Arindam",
""
],
[
"Mukherjee",
"Sumanta",
""
],
[
"Aggarwal",
"Nupur",
""
],
[
"Sarpatwar",
"Kanthi",
""
],
[
"Ganapavarapu",
"Giridhar",
""
],
[
"Vaculin",
"Roman",
""
]
] | TITLE: TsSHAP: Robust model agnostic feature-based explainability for time
series forecasting
ABSTRACT: A trustworthy machine learning model should be accurate as well as
explainable. Understanding why a model makes a certain decision defines the
notion of explainability. While various flavors of explainability have been
well-studied in supervised learning paradigms like classification and
regression, literature on explainability for time series forecasting is
relatively scarce.
In this paper, we propose a feature-based explainability algorithm, TsSHAP,
that can explain the forecast of any black-box forecasting model. The method is
agnostic of the forecasting model and can provide explanations for a forecast
in terms of interpretable features defined by the user a prior.
The explanations are in terms of the SHAP values obtained by applying the
TreeSHAP algorithm on a surrogate model that learns a mapping between the
interpretable feature space and the forecast of the black-box model.
Moreover, we formalize the notion of local, semi-local, and global
explanations in the context of time series forecasting, which can be useful in
several scenarios. We validate the efficacy and robustness of TsSHAP through
extensive experiments on multiple datasets.
| no_new_dataset | 0.940408 |
2303.12319 | Guangzheng Hu | Guangzheng Hu, Haoran Li, Shasha Liu, Mingjun Ma, Yuanheng Zhu, and
Dongbin Zhao | NeuronsMAE: A Novel Multi-Agent Reinforcement Learning Environment for
Cooperative and Competitive Multi-Robot Tasks | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Multi-agent reinforcement learning (MARL) has achieved remarkable success in
various challenging problems. Meanwhile, more and more benchmarks have emerged
and provided some standards to evaluate the algorithms in different fields. On
the one hand, the virtual MARL environments lack knowledge of real-world tasks
and actuator abilities, and on the other hand, the current task-specified
multi-robot platform has poor support for the generality of multi-agent
reinforcement learning algorithms and lacks support for transferring from
simulation to the real environment. Bridging the gap between the virtual MARL
environments and the real multi-robot platform becomes the key to promoting the
practicability of MARL algorithms. This paper proposes a novel MARL environment
for real multi-robot tasks named NeuronsMAE (Neurons Multi-Agent Environment).
This environment supports cooperative and competitive multi-robot tasks and is
configured with rich parameter interfaces to study the multi-agent policy
transfer from simulation to reality. With this platform, we evaluate various
popular MARL algorithms and build a new MARL benchmark for multi-robot tasks.
We hope that this platform will facilitate the research and application of MARL
algorithms for real robot tasks. Information about the benchmark and the
open-source code will be released.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 05:30:39 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Hu",
"Guangzheng",
""
],
[
"Li",
"Haoran",
""
],
[
"Liu",
"Shasha",
""
],
[
"Ma",
"Mingjun",
""
],
[
"Zhu",
"Yuanheng",
""
],
[
"Zhao",
"Dongbin",
""
]
] | TITLE: NeuronsMAE: A Novel Multi-Agent Reinforcement Learning Environment for
Cooperative and Competitive Multi-Robot Tasks
ABSTRACT: Multi-agent reinforcement learning (MARL) has achieved remarkable success in
various challenging problems. Meanwhile, more and more benchmarks have emerged
and provided some standards to evaluate the algorithms in different fields. On
the one hand, the virtual MARL environments lack knowledge of real-world tasks
and actuator abilities, and on the other hand, the current task-specified
multi-robot platform has poor support for the generality of multi-agent
reinforcement learning algorithms and lacks support for transferring from
simulation to the real environment. Bridging the gap between the virtual MARL
environments and the real multi-robot platform becomes the key to promoting the
practicability of MARL algorithms. This paper proposes a novel MARL environment
for real multi-robot tasks named NeuronsMAE (Neurons Multi-Agent Environment).
This environment supports cooperative and competitive multi-robot tasks and is
configured with rich parameter interfaces to study the multi-agent policy
transfer from simulation to reality. With this platform, we evaluate various
popular MARL algorithms and build a new MARL benchmark for multi-robot tasks.
We hope that this platform will facilitate the research and application of MARL
algorithms for real robot tasks. Information about the benchmark and the
open-source code will be released.
| no_new_dataset | 0.934395 |
2303.12346 | Shengming Yin | Shengming Yin, Chenfei Wu, Huan Yang, Jianfeng Wang, Xiaodong Wang,
Minheng Ni, Zhengyuan Yang, Linjie Li, Shuguang Liu, Fan Yang, Jianlong Fu,
Gong Ming, Lijuan Wang, Zicheng Liu, Houqiang Li, Nan Duan | NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose NUWA-XL, a novel Diffusion over Diffusion
architecture for eXtremely Long video generation. Most current work generates
long videos segment by segment sequentially, which normally leads to the gap
between training on short videos and inferring long videos, and the sequential
generation is inefficient. Instead, our approach adopts a ``coarse-to-fine''
process, in which the video can be generated in parallel at the same
granularity. A global diffusion model is applied to generate the keyframes
across the entire time range, and then local diffusion models recursively fill
in the content between nearby frames. This simple yet effective strategy allows
us to directly train on long videos (3376 frames) to reduce the
training-inference gap, and makes it possible to generate all segments in
parallel. To evaluate our model, we build FlintstonesHD dataset, a new
benchmark for long video generation. Experiments show that our model not only
generates high-quality long videos with both global and local coherence, but
also decreases the average inference time from 7.55min to 26s (by 94.26\%) at
the same hardware setting when generating 1024 frames. The homepage link is
\url{https://msra-nuwa.azurewebsites.net/}
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 07:10:09 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Yin",
"Shengming",
""
],
[
"Wu",
"Chenfei",
""
],
[
"Yang",
"Huan",
""
],
[
"Wang",
"Jianfeng",
""
],
[
"Wang",
"Xiaodong",
""
],
[
"Ni",
"Minheng",
""
],
[
"Yang",
"Zhengyuan",
""
],
[
"Li",
"Linjie",
""
],
[
"Liu",
"Shuguang",
""
],
[
"Yang",
"Fan",
""
],
[
"Fu",
"Jianlong",
""
],
[
"Ming",
"Gong",
""
],
[
"Wang",
"Lijuan",
""
],
[
"Liu",
"Zicheng",
""
],
[
"Li",
"Houqiang",
""
],
[
"Duan",
"Nan",
""
]
] | TITLE: NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation
ABSTRACT: In this paper, we propose NUWA-XL, a novel Diffusion over Diffusion
architecture for eXtremely Long video generation. Most current work generates
long videos segment by segment sequentially, which normally leads to the gap
between training on short videos and inferring long videos, and the sequential
generation is inefficient. Instead, our approach adopts a ``coarse-to-fine''
process, in which the video can be generated in parallel at the same
granularity. A global diffusion model is applied to generate the keyframes
across the entire time range, and then local diffusion models recursively fill
in the content between nearby frames. This simple yet effective strategy allows
us to directly train on long videos (3376 frames) to reduce the
training-inference gap, and makes it possible to generate all segments in
parallel. To evaluate our model, we build FlintstonesHD dataset, a new
benchmark for long video generation. Experiments show that our model not only
generates high-quality long videos with both global and local coherence, but
also decreases the average inference time from 7.55min to 26s (by 94.26\%) at
the same hardware setting when generating 1024 frames. The homepage link is
\url{https://msra-nuwa.azurewebsites.net/}
| new_dataset | 0.953622 |
2303.12352 | Frances Fengyi Yang | Frances Fengyi Yang and Michele Sasdelli and Tat-Jun Chin | Training Multilayer Perceptrons by Sampling with Quantum Annealers | 22 pages, 15 figures | null | null | null | cs.LG quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A successful application of quantum annealing to machine learning is training
restricted Boltzmann machines (RBM). However, many neural networks for vision
applications are feedforward structures, such as multilayer perceptrons (MLP).
Backpropagation is currently the most effective technique to train MLPs for
supervised learning. This paper aims to be forward-looking by exploring the
training of MLPs using quantum annealers. We exploit an equivalence between
MLPs and energy-based models (EBM), which are a variation of RBMs with a
maximum conditional likelihood objective. This leads to a strategy to train
MLPs with quantum annealers as a sampling engine. We prove our setup for MLPs
with sigmoid activation functions and one hidden layer, and demonstrated
training of binary image classifiers on small subsets of the MNIST and
Fashion-MNIST datasets using the D-Wave quantum annealer. Although problem
sizes that are feasible on current annealers are limited, we obtained
comprehensive results on feasible instances that validate our ideas. Our work
establishes the potential of quantum computing for training MLPs.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 07:40:01 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Yang",
"Frances Fengyi",
""
],
[
"Sasdelli",
"Michele",
""
],
[
"Chin",
"Tat-Jun",
""
]
] | TITLE: Training Multilayer Perceptrons by Sampling with Quantum Annealers
ABSTRACT: A successful application of quantum annealing to machine learning is training
restricted Boltzmann machines (RBM). However, many neural networks for vision
applications are feedforward structures, such as multilayer perceptrons (MLP).
Backpropagation is currently the most effective technique to train MLPs for
supervised learning. This paper aims to be forward-looking by exploring the
training of MLPs using quantum annealers. We exploit an equivalence between
MLPs and energy-based models (EBM), which are a variation of RBMs with a
maximum conditional likelihood objective. This leads to a strategy to train
MLPs with quantum annealers as a sampling engine. We prove our setup for MLPs
with sigmoid activation functions and one hidden layer, and demonstrated
training of binary image classifiers on small subsets of the MNIST and
Fashion-MNIST datasets using the D-Wave quantum annealer. Although problem
sizes that are feasible on current annealers are limited, we obtained
comprehensive results on feasible instances that validate our ideas. Our work
establishes the potential of quantum computing for training MLPs.
| no_new_dataset | 0.948298 |
2303.12357 | Wenjie Wang | Wenjie Wang, Li Xiong, Jian Lou | Wasserstein Adversarial Examples on Univariant Time Series Data | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial examples are crafted by adding indistinguishable perturbations to
normal examples in order to fool a well-trained deep learning model to
misclassify. In the context of computer vision, this notion of
indistinguishability is typically bounded by $L_{\infty}$ or other norms.
However, these norms are not appropriate for measuring indistinguishiability
for time series data. In this work, we propose adversarial examples in the
Wasserstein space for time series data for the first time and utilize
Wasserstein distance to bound the perturbation between normal examples and
adversarial examples. We introduce Wasserstein projected gradient descent
(WPGD), an adversarial attack method for perturbing univariant time series
data. We leverage the closed-form solution of Wasserstein distance in the 1D
space to calculate the projection step of WPGD efficiently with the gradient
descent method. We further propose a two-step projection so that the search of
adversarial examples in the Wasserstein space is guided and constrained by
Euclidean norms to yield more effective and imperceptible perturbations. We
empirically evaluate the proposed attack on several time series datasets in the
healthcare domain. Extensive results demonstrate that the Wasserstein attack is
powerful and can successfully attack most of the target classifiers with a high
attack success rate. To better study the nature of Wasserstein adversarial
example, we evaluate a strong defense mechanism named Wasserstein smoothing for
potential certified robustness defense. Although the defense can achieve some
accuracy gain, it still has limitations in many cases and leaves space for
developing a stronger certified robustness method to Wasserstein adversarial
examples on univariant time series data.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 07:50:15 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wang",
"Wenjie",
""
],
[
"Xiong",
"Li",
""
],
[
"Lou",
"Jian",
""
]
] | TITLE: Wasserstein Adversarial Examples on Univariant Time Series Data
ABSTRACT: Adversarial examples are crafted by adding indistinguishable perturbations to
normal examples in order to fool a well-trained deep learning model to
misclassify. In the context of computer vision, this notion of
indistinguishability is typically bounded by $L_{\infty}$ or other norms.
However, these norms are not appropriate for measuring indistinguishiability
for time series data. In this work, we propose adversarial examples in the
Wasserstein space for time series data for the first time and utilize
Wasserstein distance to bound the perturbation between normal examples and
adversarial examples. We introduce Wasserstein projected gradient descent
(WPGD), an adversarial attack method for perturbing univariant time series
data. We leverage the closed-form solution of Wasserstein distance in the 1D
space to calculate the projection step of WPGD efficiently with the gradient
descent method. We further propose a two-step projection so that the search of
adversarial examples in the Wasserstein space is guided and constrained by
Euclidean norms to yield more effective and imperceptible perturbations. We
empirically evaluate the proposed attack on several time series datasets in the
healthcare domain. Extensive results demonstrate that the Wasserstein attack is
powerful and can successfully attack most of the target classifiers with a high
attack success rate. To better study the nature of Wasserstein adversarial
example, we evaluate a strong defense mechanism named Wasserstein smoothing for
potential certified robustness defense. Although the defense can achieve some
accuracy gain, it still has limitations in many cases and leaves space for
developing a stronger certified robustness method to Wasserstein adversarial
examples on univariant time series data.
| no_new_dataset | 0.945197 |
2303.12369 | Hui Lv | Hui Lv, Zhongqi Yue, Qianru Sun, Bin Luo, Zhen Cui, Hanwang Zhang | Unbiased Multiple Instance Learning for Weakly Supervised Video Anomaly
Detection | 11 pages,10 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Weakly Supervised Video Anomaly Detection (WSVAD) is challenging because the
binary anomaly label is only given on the video level, but the output requires
snippet-level predictions. So, Multiple Instance Learning (MIL) is prevailing
in WSVAD. However, MIL is notoriously known to suffer from many false alarms
because the snippet-level detector is easily biased towards the abnormal
snippets with simple context, confused by the normality with the same bias, and
missing the anomaly with a different pattern. To this end, we propose a new MIL
framework: Unbiased MIL (UMIL), to learn unbiased anomaly features that improve
WSVAD. At each MIL training iteration, we use the current detector to divide
the samples into two groups with different context biases: the most confident
abnormal/normal snippets and the rest ambiguous ones. Then, by seeking the
invariant features across the two sample groups, we can remove the variant
context biases. Extensive experiments on benchmarks UCF-Crime and TAD
demonstrate the effectiveness of our UMIL. Our code is provided at
https://github.com/ktr-hubrt/UMIL.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 08:11:22 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Lv",
"Hui",
""
],
[
"Yue",
"Zhongqi",
""
],
[
"Sun",
"Qianru",
""
],
[
"Luo",
"Bin",
""
],
[
"Cui",
"Zhen",
""
],
[
"Zhang",
"Hanwang",
""
]
] | TITLE: Unbiased Multiple Instance Learning for Weakly Supervised Video Anomaly
Detection
ABSTRACT: Weakly Supervised Video Anomaly Detection (WSVAD) is challenging because the
binary anomaly label is only given on the video level, but the output requires
snippet-level predictions. So, Multiple Instance Learning (MIL) is prevailing
in WSVAD. However, MIL is notoriously known to suffer from many false alarms
because the snippet-level detector is easily biased towards the abnormal
snippets with simple context, confused by the normality with the same bias, and
missing the anomaly with a different pattern. To this end, we propose a new MIL
framework: Unbiased MIL (UMIL), to learn unbiased anomaly features that improve
WSVAD. At each MIL training iteration, we use the current detector to divide
the samples into two groups with different context biases: the most confident
abnormal/normal snippets and the rest ambiguous ones. Then, by seeking the
invariant features across the two sample groups, we can remove the variant
context biases. Extensive experiments on benchmarks UCF-Crime and TAD
demonstrate the effectiveness of our UMIL. Our code is provided at
https://github.com/ktr-hubrt/UMIL.
| no_new_dataset | 0.947137 |
2303.12379 | Yi-Shan Lee | Yi-Shan Lee, Wei-Cheng Tseng, Fu-En Wang, Min Sun | VMCML: Video and Music Matching via Cross-Modality Lifting | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a content-based system for matching video and background music.
The system aims to address the challenges in music recommendation for new users
or new music give short-form videos. To this end, we propose a cross-modal
framework VMCML that finds a shared embedding space between video and music
representations. To ensure the embedding space can be effectively shared by
both representations, we leverage CosFace loss based on margin-based cosine
similarity loss. Furthermore, we establish a large-scale dataset called MSVD,
in which we provide 390 individual music and the corresponding matched 150,000
videos. We conduct extensive experiments on Youtube-8M and our MSVD datasets.
Our quantitative and qualitative results demonstrate the effectiveness of our
proposed framework and achieve state-of-the-art video and music matching
performance.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 08:28:23 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Lee",
"Yi-Shan",
""
],
[
"Tseng",
"Wei-Cheng",
""
],
[
"Wang",
"Fu-En",
""
],
[
"Sun",
"Min",
""
]
] | TITLE: VMCML: Video and Music Matching via Cross-Modality Lifting
ABSTRACT: We propose a content-based system for matching video and background music.
The system aims to address the challenges in music recommendation for new users
or new music give short-form videos. To this end, we propose a cross-modal
framework VMCML that finds a shared embedding space between video and music
representations. To ensure the embedding space can be effectively shared by
both representations, we leverage CosFace loss based on margin-based cosine
similarity loss. Furthermore, we establish a large-scale dataset called MSVD,
in which we provide 390 individual music and the corresponding matched 150,000
videos. We conduct extensive experiments on Youtube-8M and our MSVD datasets.
Our quantitative and qualitative results demonstrate the effectiveness of our
proposed framework and achieve state-of-the-art video and music matching
performance.
| new_dataset | 0.955236 |
2303.12396 | Yang Hai | Yang Hai, Rui Song, Jiaojiao Li, Mathieu Salzmann, Yinlin Hu | Rigidity-Aware Detection for 6D Object Pose Estimation | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most recent 6D object pose estimation methods first use object detection to
obtain 2D bounding boxes before actually regressing the pose. However, the
general object detection methods they use are ill-suited to handle cluttered
scenes, thus producing poor initialization to the subsequent pose network. To
address this, we propose a rigidity-aware detection method exploiting the fact
that, in 6D pose estimation, the target objects are rigid. This lets us
introduce an approach to sampling positive object regions from the entire
visible object area during training, instead of naively drawing samples from
the bounding box center where the object might be occluded. As such, every
visible object part can contribute to the final bounding box prediction,
yielding better detection robustness. Key to the success of our approach is a
visibility map, which we propose to build using a minimum barrier distance
between every pixel in the bounding box and the box boundary. Our results on
seven challenging 6D pose estimation datasets evidence that our method
outperforms general detection frameworks by a large margin. Furthermore,
combined with a pose regression network, we obtain state-of-the-art pose
estimation results on the challenging BOP benchmark.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 09:02:54 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Hai",
"Yang",
""
],
[
"Song",
"Rui",
""
],
[
"Li",
"Jiaojiao",
""
],
[
"Salzmann",
"Mathieu",
""
],
[
"Hu",
"Yinlin",
""
]
] | TITLE: Rigidity-Aware Detection for 6D Object Pose Estimation
ABSTRACT: Most recent 6D object pose estimation methods first use object detection to
obtain 2D bounding boxes before actually regressing the pose. However, the
general object detection methods they use are ill-suited to handle cluttered
scenes, thus producing poor initialization to the subsequent pose network. To
address this, we propose a rigidity-aware detection method exploiting the fact
that, in 6D pose estimation, the target objects are rigid. This lets us
introduce an approach to sampling positive object regions from the entire
visible object area during training, instead of naively drawing samples from
the bounding box center where the object might be occluded. As such, every
visible object part can contribute to the final bounding box prediction,
yielding better detection robustness. Key to the success of our approach is a
visibility map, which we propose to build using a minimum barrier distance
between every pixel in the bounding box and the box boundary. Our results on
seven challenging 6D pose estimation datasets evidence that our method
outperforms general detection frameworks by a large margin. Furthermore,
combined with a pose regression network, we obtain state-of-the-art pose
estimation results on the challenging BOP benchmark.
| no_new_dataset | 0.942348 |
2303.12400 | Tianhang Wang | Tianhang Wang, Guang Chen, Kai Chen, Zhengfa Liu, Bo Zhang, Alois
Knoll, Changjun Jiang | UMC: A Unified Bandwidth-efficient and Multi-resolution based
Collaborative Perception Framework | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-agent collaborative perception (MCP) has recently attracted much
attention. It includes three key processes: communication for sharing,
collaboration for integration, and reconstruction for different downstream
tasks. Existing methods pursue designing the collaboration process alone,
ignoring their intrinsic interactions and resulting in suboptimal performance.
In contrast, we aim to propose a Unified Collaborative perception framework
named UMC, optimizing the communication, collaboration, and reconstruction
processes with the Multi-resolution technique. The communication introduces a
novel trainable multi-resolution and selective-region (MRSR) mechanism,
achieving higher quality and lower bandwidth. Then, a graph-based collaboration
is proposed, conducting on each resolution to adapt the MRSR. Finally, the
reconstruction integrates the multi-resolution collaborative features for
downstream tasks. Since the general metric can not reflect the performance
enhancement brought by MCP systematically, we introduce a brand-new evaluation
metric that evaluates the MCP from different perspectives. To verify our
algorithm, we conducted experiments on the V2X-Sim and OPV2V datasets. Our
quantitative and qualitative experiments prove that the proposed UMC greatly
outperforms the state-of-the-art collaborative perception approaches.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 09:09:02 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wang",
"Tianhang",
""
],
[
"Chen",
"Guang",
""
],
[
"Chen",
"Kai",
""
],
[
"Liu",
"Zhengfa",
""
],
[
"Zhang",
"Bo",
""
],
[
"Knoll",
"Alois",
""
],
[
"Jiang",
"Changjun",
""
]
] | TITLE: UMC: A Unified Bandwidth-efficient and Multi-resolution based
Collaborative Perception Framework
ABSTRACT: Multi-agent collaborative perception (MCP) has recently attracted much
attention. It includes three key processes: communication for sharing,
collaboration for integration, and reconstruction for different downstream
tasks. Existing methods pursue designing the collaboration process alone,
ignoring their intrinsic interactions and resulting in suboptimal performance.
In contrast, we aim to propose a Unified Collaborative perception framework
named UMC, optimizing the communication, collaboration, and reconstruction
processes with the Multi-resolution technique. The communication introduces a
novel trainable multi-resolution and selective-region (MRSR) mechanism,
achieving higher quality and lower bandwidth. Then, a graph-based collaboration
is proposed, conducting on each resolution to adapt the MRSR. Finally, the
reconstruction integrates the multi-resolution collaborative features for
downstream tasks. Since the general metric can not reflect the performance
enhancement brought by MCP systematically, we introduce a brand-new evaluation
metric that evaluates the MCP from different perspectives. To verify our
algorithm, we conducted experiments on the V2X-Sim and OPV2V datasets. Our
quantitative and qualitative experiments prove that the proposed UMC greatly
outperforms the state-of-the-art collaborative perception approaches.
| no_new_dataset | 0.945851 |
2303.12489 | Aman Chadha Mr. | Aman Chadha, Vinija Jain | Few-shot Multimodal Multitask Multilingual Learning | null | null | null | null | cs.LG cs.AI cs.CL cs.CV cs.MM | http://creativecommons.org/licenses/by/4.0/ | While few-shot learning as a transfer learning paradigm has gained
significant traction for scenarios with limited data, it has primarily been
explored in the context of building unimodal and unilingual models.
Furthermore, a significant part of the existing literature in the domain of
few-shot multitask learning perform in-context learning which requires manually
generated prompts as the input, yielding varying outcomes depending on the
level of manual prompt-engineering. In addition, in-context learning suffers
from substantial computational, memory, and storage costs which eventually
leads to high inference latency because it involves running all of the prompt's
examples through the model every time a prediction is made. In contrast,
methods based on the transfer learning via the fine-tuning paradigm avoid the
aforementioned issues at a one-time cost of fine-tuning weights on a per-task
basis. However, such methods lack exposure to few-shot multimodal multitask
learning. In this paper, we propose few-shot learning for a multimodal
multitask multilingual (FM3) setting by adapting pre-trained vision and
language models using task-specific hypernetworks and contrastively fine-tuning
them to enable few-shot learning. FM3's architecture combines the best of both
worlds of in-context and fine-tuning based learning and consists of three major
components: (i) multimodal contrastive fine-tuning to enable few-shot learning,
(ii) hypernetwork task adaptation to perform multitask learning, and (iii)
task-specific output heads to cater to a plethora of diverse tasks. FM3 learns
the most prominent tasks in the vision and language domains along with their
intersections, namely visual entailment (VE), visual question answering (VQA),
and natural language understanding (NLU) tasks such as neural entity
recognition (NER) and the GLUE benchmark including QNLI, MNLI, QQP, and SST-2.
| [
{
"version": "v1",
"created": "Sun, 19 Feb 2023 03:48:46 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Chadha",
"Aman",
""
],
[
"Jain",
"Vinija",
""
]
] | TITLE: Few-shot Multimodal Multitask Multilingual Learning
ABSTRACT: While few-shot learning as a transfer learning paradigm has gained
significant traction for scenarios with limited data, it has primarily been
explored in the context of building unimodal and unilingual models.
Furthermore, a significant part of the existing literature in the domain of
few-shot multitask learning perform in-context learning which requires manually
generated prompts as the input, yielding varying outcomes depending on the
level of manual prompt-engineering. In addition, in-context learning suffers
from substantial computational, memory, and storage costs which eventually
leads to high inference latency because it involves running all of the prompt's
examples through the model every time a prediction is made. In contrast,
methods based on the transfer learning via the fine-tuning paradigm avoid the
aforementioned issues at a one-time cost of fine-tuning weights on a per-task
basis. However, such methods lack exposure to few-shot multimodal multitask
learning. In this paper, we propose few-shot learning for a multimodal
multitask multilingual (FM3) setting by adapting pre-trained vision and
language models using task-specific hypernetworks and contrastively fine-tuning
them to enable few-shot learning. FM3's architecture combines the best of both
worlds of in-context and fine-tuning based learning and consists of three major
components: (i) multimodal contrastive fine-tuning to enable few-shot learning,
(ii) hypernetwork task adaptation to perform multitask learning, and (iii)
task-specific output heads to cater to a plethora of diverse tasks. FM3 learns
the most prominent tasks in the vision and language domains along with their
intersections, namely visual entailment (VE), visual question answering (VQA),
and natural language understanding (NLU) tasks such as neural entity
recognition (NER) and the GLUE benchmark including QNLI, MNLI, QQP, and SST-2.
| no_new_dataset | 0.950273 |
2303.12492 | Vincenzo Suriani | Vincenzo Suriani, Daniele Nardi | Preserving HRI Capabilities: Physical, Remote and Simulated Modalities
in the SciRoc 2021 Competition | HRI 2023 Workshop on Advancing HRI Researchand Benchmarking Through
Open-Source Ecosystems, March 13, 2023, Stockholm, Sweden | null | null | null | cs.RO cs.HC | http://creativecommons.org/licenses/by/4.0/ | In the last years, robots are moving out of research laboratories to enter
everyday life. Competitions aiming at benchmarking the capabilities of a robot
in everyday scenarios are useful to make a step forward in this path. In fact,
they foster the development of robust architectures capable of solving issues
that might occur during human-robot coexistence in human-shaped scenarios. One
of those competitions is SciRoc that, in its second edition, proposed new
benchmarking environments. In particular, Episode 1 of SciRoc 2 proposed three
different modalities of participation while preserving the Human-Robot
Interaction (HRI), being a fundamental benchmarking functionality. The Coffee
Shop environment, used to challenge the participating teams, represented an
excellent testbed enabling for the benchmarking of different robotics
functionalities, but also an exceptional opportunity for proposing novel
solutions to guarantee real human-robot interaction procedures despite the
Covid-19 pandemic restrictions. The developed software is publicly released.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 12:05:25 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Suriani",
"Vincenzo",
""
],
[
"Nardi",
"Daniele",
""
]
] | TITLE: Preserving HRI Capabilities: Physical, Remote and Simulated Modalities
in the SciRoc 2021 Competition
ABSTRACT: In the last years, robots are moving out of research laboratories to enter
everyday life. Competitions aiming at benchmarking the capabilities of a robot
in everyday scenarios are useful to make a step forward in this path. In fact,
they foster the development of robust architectures capable of solving issues
that might occur during human-robot coexistence in human-shaped scenarios. One
of those competitions is SciRoc that, in its second edition, proposed new
benchmarking environments. In particular, Episode 1 of SciRoc 2 proposed three
different modalities of participation while preserving the Human-Robot
Interaction (HRI), being a fundamental benchmarking functionality. The Coffee
Shop environment, used to challenge the participating teams, represented an
excellent testbed enabling for the benchmarking of different robotics
functionalities, but also an exceptional opportunity for proposing novel
solutions to guarantee real human-robot interaction procedures despite the
Covid-19 pandemic restrictions. The developed software is publicly released.
| no_new_dataset | 0.943086 |
2303.12501 | Ding Jiang | Ding Jiang, Mang Ye | Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image
Person Retrieval | Accepted by CVPR 2023. Codes are available at this
https://github.com/anosorae/IRRA | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text-to-image person retrieval aims to identify the target person based on a
given textual description query. The primary challenge is to learn the mapping
of visual and textual modalities into a common latent space. Prior works have
attempted to address this challenge by leveraging separately pre-trained
unimodal models to extract visual and textual features. However, these
approaches lack the necessary underlying alignment capabilities required to
match multimodal data effectively. Besides, these works use prior information
to explore explicit part alignments, which may lead to the distortion of
intra-modality information. To alleviate these issues, we present IRRA: a
cross-modal Implicit Relation Reasoning and Aligning framework that learns
relations between local visual-textual tokens and enhances global image-text
matching without requiring additional prior supervision. Specifically, we first
design an Implicit Relation Reasoning module in a masked language modeling
paradigm. This achieves cross-modal interaction by integrating the visual cues
into the textual tokens with a cross-modal multimodal interaction encoder.
Secondly, to globally align the visual and textual embeddings, Similarity
Distribution Matching is proposed to minimize the KL divergence between
image-text similarity distributions and the normalized label matching
distributions. The proposed method achieves new state-of-the-art results on all
three public datasets, with a notable margin of about 3%-9% for Rank-1 accuracy
compared to prior methods.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 12:11:59 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Jiang",
"Ding",
""
],
[
"Ye",
"Mang",
""
]
] | TITLE: Cross-Modal Implicit Relation Reasoning and Aligning for Text-to-Image
Person Retrieval
ABSTRACT: Text-to-image person retrieval aims to identify the target person based on a
given textual description query. The primary challenge is to learn the mapping
of visual and textual modalities into a common latent space. Prior works have
attempted to address this challenge by leveraging separately pre-trained
unimodal models to extract visual and textual features. However, these
approaches lack the necessary underlying alignment capabilities required to
match multimodal data effectively. Besides, these works use prior information
to explore explicit part alignments, which may lead to the distortion of
intra-modality information. To alleviate these issues, we present IRRA: a
cross-modal Implicit Relation Reasoning and Aligning framework that learns
relations between local visual-textual tokens and enhances global image-text
matching without requiring additional prior supervision. Specifically, we first
design an Implicit Relation Reasoning module in a masked language modeling
paradigm. This achieves cross-modal interaction by integrating the visual cues
into the textual tokens with a cross-modal multimodal interaction encoder.
Secondly, to globally align the visual and textual embeddings, Similarity
Distribution Matching is proposed to minimize the KL divergence between
image-text similarity distributions and the normalized label matching
distributions. The proposed method achieves new state-of-the-art results on all
three public datasets, with a notable margin of about 3%-9% for Rank-1 accuracy
compared to prior methods.
| no_new_dataset | 0.94625 |
2303.12507 | Yan Luo | Yan Luo, Ye Liu, Fu-lai Chung, Yu Liu, Chang Wen Chen | End-to-End Personalized Next Location Recommendation via Contrastive
User Preference Modeling | null | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | Predicting the next location is a highly valuable and common need in many
location-based services such as destination prediction and route planning. The
goal of next location recommendation is to predict the next point-of-interest a
user might go to based on the user's historical trajectory. Most existing
models learn mobility patterns merely from users' historical check-in sequences
while overlooking the significance of user preference modeling. In this work, a
novel Point-of-Interest Transformer (POIFormer) with contrastive user
preference modeling is developed for end-to-end next location recommendation.
This model consists of three major modules: history encoder, query generator,
and preference decoder. History encoder is designed to model mobility patterns
from historical check-in sequences, while query generator explicitly learns
user preferences to generate user-specific intention queries. Finally,
preference decoder combines the intention queries and historical information to
predict the user's next location. Extensive comparisons with representative
schemes and ablation studies on four real-world datasets demonstrate the
effectiveness and superiority of the proposed scheme under various settings.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 12:29:18 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Luo",
"Yan",
""
],
[
"Liu",
"Ye",
""
],
[
"Chung",
"Fu-lai",
""
],
[
"Liu",
"Yu",
""
],
[
"Chen",
"Chang Wen",
""
]
] | TITLE: End-to-End Personalized Next Location Recommendation via Contrastive
User Preference Modeling
ABSTRACT: Predicting the next location is a highly valuable and common need in many
location-based services such as destination prediction and route planning. The
goal of next location recommendation is to predict the next point-of-interest a
user might go to based on the user's historical trajectory. Most existing
models learn mobility patterns merely from users' historical check-in sequences
while overlooking the significance of user preference modeling. In this work, a
novel Point-of-Interest Transformer (POIFormer) with contrastive user
preference modeling is developed for end-to-end next location recommendation.
This model consists of three major modules: history encoder, query generator,
and preference decoder. History encoder is designed to model mobility patterns
from historical check-in sequences, while query generator explicitly learns
user preferences to generate user-specific intention queries. Finally,
preference decoder combines the intention queries and historical information to
predict the user's next location. Extensive comparisons with representative
schemes and ablation studies on four real-world datasets demonstrate the
effectiveness and superiority of the proposed scheme under various settings.
| no_new_dataset | 0.950088 |
2303.12589 | Naoto Yanai | Yumeki Goto and Nami Ashizawa and Toshiki Shibahara and Naoto Yanai | Do Backdoors Assist Membership Inference Attacks? | null | null | null | null | cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | When an adversary provides poison samples to a machine learning model,
privacy leakage, such as membership inference attacks that infer whether a
sample was included in the training of the model, becomes effective by moving
the sample to an outlier. However, the attacks can be detected because
inference accuracy deteriorates due to poison samples. In this paper, we
discuss a \textit{backdoor-assisted membership inference attack}, a novel
membership inference attack based on backdoors that return the adversary's
expected output for a triggered sample. We found three crucial insights through
experiments with an academic benchmark dataset. We first demonstrate that the
backdoor-assisted membership inference attack is unsuccessful. Second, when we
analyzed loss distributions to understand the reason for the unsuccessful
results, we found that backdoors cannot separate loss distributions of training
and non-training samples. In other words, backdoors cannot affect the
distribution of clean samples. Third, we also show that poison and triggered
samples activate neurons of different distributions. Specifically, backdoors
make any clean sample an inlier, contrary to poisoning samples. As a result, we
confirm that backdoors cannot assist membership inference.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 14:19:06 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Goto",
"Yumeki",
""
],
[
"Ashizawa",
"Nami",
""
],
[
"Shibahara",
"Toshiki",
""
],
[
"Yanai",
"Naoto",
""
]
] | TITLE: Do Backdoors Assist Membership Inference Attacks?
ABSTRACT: When an adversary provides poison samples to a machine learning model,
privacy leakage, such as membership inference attacks that infer whether a
sample was included in the training of the model, becomes effective by moving
the sample to an outlier. However, the attacks can be detected because
inference accuracy deteriorates due to poison samples. In this paper, we
discuss a \textit{backdoor-assisted membership inference attack}, a novel
membership inference attack based on backdoors that return the adversary's
expected output for a triggered sample. We found three crucial insights through
experiments with an academic benchmark dataset. We first demonstrate that the
backdoor-assisted membership inference attack is unsuccessful. Second, when we
analyzed loss distributions to understand the reason for the unsuccessful
results, we found that backdoors cannot separate loss distributions of training
and non-training samples. In other words, backdoors cannot affect the
distribution of clean samples. Third, we also show that poison and triggered
samples activate neurons of different distributions. Specifically, backdoors
make any clean sample an inlier, contrary to poisoning samples. As a result, we
confirm that backdoors cannot assist membership inference.
| no_new_dataset | 0.946794 |
2303.12618 | Zicheng Zhang | Zicheng Zhang, Chunyi Li, Wei Sun, Xiaohong Liu, Xiongkuo Min,
Guangtao Zhai | A Perceptual Quality Assessment Exploration for AIGC Images | null | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | \underline{AI} \underline{G}enerated \underline{C}ontent (\textbf{AIGC}) has
gained widespread attention with the increasing efficiency of deep learning in
content creation. AIGC, created with the assistance of artificial intelligence
technology, includes various forms of content, among which the AI-generated
images (AGIs) have brought significant impact to society and have been applied
to various fields such as entertainment, education, social media, etc. However,
due to hardware limitations and technical proficiency, the quality of AIGC
images (AGIs) varies, necessitating refinement and filtering before practical
use. Consequently, there is an urgent need for developing objective models to
assess the quality of AGIs. Unfortunately, no research has been carried out to
investigate the perceptual quality assessment for AGIs specifically. Therefore,
in this paper, we first discuss the major evaluation aspects such as technical
issues, AI artifacts, unnaturalness, discrepancy, and aesthetics for AGI
quality assessment. Then we present the first perceptual AGI quality assessment
database, AGIQA-1K, which consists of 1,080 AGIs generated from diffusion
models. A well-organized subjective experiment is followed to collect the
quality labels of the AGIs. Finally, we conduct a benchmark experiment to
evaluate the performance of current image quality assessment (IQA) models.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 14:59:49 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhang",
"Zicheng",
""
],
[
"Li",
"Chunyi",
""
],
[
"Sun",
"Wei",
""
],
[
"Liu",
"Xiaohong",
""
],
[
"Min",
"Xiongkuo",
""
],
[
"Zhai",
"Guangtao",
""
]
] | TITLE: A Perceptual Quality Assessment Exploration for AIGC Images
ABSTRACT: \underline{AI} \underline{G}enerated \underline{C}ontent (\textbf{AIGC}) has
gained widespread attention with the increasing efficiency of deep learning in
content creation. AIGC, created with the assistance of artificial intelligence
technology, includes various forms of content, among which the AI-generated
images (AGIs) have brought significant impact to society and have been applied
to various fields such as entertainment, education, social media, etc. However,
due to hardware limitations and technical proficiency, the quality of AIGC
images (AGIs) varies, necessitating refinement and filtering before practical
use. Consequently, there is an urgent need for developing objective models to
assess the quality of AGIs. Unfortunately, no research has been carried out to
investigate the perceptual quality assessment for AGIs specifically. Therefore,
in this paper, we first discuss the major evaluation aspects such as technical
issues, AI artifacts, unnaturalness, discrepancy, and aesthetics for AGI
quality assessment. Then we present the first perceptual AGI quality assessment
database, AGIQA-1K, which consists of 1,080 AGIs generated from diffusion
models. A well-organized subjective experiment is followed to collect the
quality labels of the AGIs. Finally, we conduct a benchmark experiment to
evaluate the performance of current image quality assessment (IQA) models.
| no_new_dataset | 0.742468 |
2303.12634 | Sumanta Mukherjee | Shravan Kumar Sajja, Sumanta Mukherjee, Satyam Dwivedi | Semi-supervised counterfactual explanations | 12 pages, 3 figures, 4 tables | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Counterfactual explanations for machine learning models are used to find
minimal interventions to the feature values such that the model changes the
prediction to a different output or a target output. A valid counterfactual
explanation should have likely feature values. Here, we address the challenge
of generating counterfactual explanations that lie in the same data
distribution as that of the training data and more importantly, they belong to
the target class distribution. This requirement has been addressed through the
incorporation of auto-encoder reconstruction loss in the counterfactual search
process. Connecting the output behavior of the classifier to the latent space
of the auto-encoder has further improved the speed of the counterfactual search
process and the interpretability of the resulting counterfactual explanations.
Continuing this line of research, we show further improvement in the
interpretability of counterfactual explanations when the auto-encoder is
trained in a semi-supervised fashion with class tagged input data. We
empirically evaluate our approach on several datasets and show considerable
improvement in-terms of several metrics.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 15:17:16 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Sajja",
"Shravan Kumar",
""
],
[
"Mukherjee",
"Sumanta",
""
],
[
"Dwivedi",
"Satyam",
""
]
] | TITLE: Semi-supervised counterfactual explanations
ABSTRACT: Counterfactual explanations for machine learning models are used to find
minimal interventions to the feature values such that the model changes the
prediction to a different output or a target output. A valid counterfactual
explanation should have likely feature values. Here, we address the challenge
of generating counterfactual explanations that lie in the same data
distribution as that of the training data and more importantly, they belong to
the target class distribution. This requirement has been addressed through the
incorporation of auto-encoder reconstruction loss in the counterfactual search
process. Connecting the output behavior of the classifier to the latent space
of the auto-encoder has further improved the speed of the counterfactual search
process and the interpretability of the resulting counterfactual explanations.
Continuing this line of research, we show further improvement in the
interpretability of counterfactual explanations when the auto-encoder is
trained in a semi-supervised fashion with class tagged input data. We
empirically evaluate our approach on several datasets and show considerable
improvement in-terms of several metrics.
| no_new_dataset | 0.94699 |
2303.12643 | Lokesh Chandra Das | Lokesh Chandra Das | Traffic Volume Prediction using Memory-Based Recurrent Neural Networks:
A comparative analysis of LSTM and GRU | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Predicting traffic volume in real-time can improve both traffic flow and road
safety. A precise traffic volume forecast helps alert drivers to the flow of
traffic along their preferred routes, preventing potential deadlock situations.
Existing parametric models cannot reliably forecast traffic volume in dynamic
and complex traffic conditions. Therefore, in order to evaluate and forecast
the traffic volume for every given time step in a real-time manner, we develop
non-linear memory-based deep neural network models. Our extensive experiments
run on the Metro Interstate Traffic Volume dataset demonstrate the
effectiveness of the proposed models in predicting traffic volume in highly
dynamic and heterogeneous traffic environments.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 15:25:07 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Das",
"Lokesh Chandra",
""
]
] | TITLE: Traffic Volume Prediction using Memory-Based Recurrent Neural Networks:
A comparative analysis of LSTM and GRU
ABSTRACT: Predicting traffic volume in real-time can improve both traffic flow and road
safety. A precise traffic volume forecast helps alert drivers to the flow of
traffic along their preferred routes, preventing potential deadlock situations.
Existing parametric models cannot reliably forecast traffic volume in dynamic
and complex traffic conditions. Therefore, in order to evaluate and forecast
the traffic volume for every given time step in a real-time manner, we develop
non-linear memory-based deep neural network models. Our extensive experiments
run on the Metro Interstate Traffic Volume dataset demonstrate the
effectiveness of the proposed models in predicting traffic volume in highly
dynamic and heterogeneous traffic environments.
| no_new_dataset | 0.945601 |
2303.12658 | Xunguang Wang | Xunguang Wang, Jiawang Bai, Xinyue Xu, Xiaomeng Li | Reliable and Efficient Evaluation of Adversarial Robustness for Deep
Hashing-Based Retrieval | arXiv admin note: text overlap with arXiv:2204.10779 | null | null | null | cs.CV cs.AI cs.CR cs.LG cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep hashing has been extensively applied to massive image retrieval due to
its efficiency and effectiveness. Recently, several adversarial attacks have
been presented to reveal the vulnerability of deep hashing models against
adversarial examples. However, existing attack methods suffer from degraded
performance or inefficiency because they underutilize the semantic relations
between original samples or spend a lot of time learning these relations with a
deep neural network. In this paper, we propose a novel Pharos-guided Attack,
dubbed PgA, to evaluate the adversarial robustness of deep hashing networks
reliably and efficiently. Specifically, we design pharos code to represent the
semantics of the benign image, which preserves the similarity to semantically
relevant samples and dissimilarity to irrelevant ones. It is proven that we can
quickly calculate the pharos code via a simple math formula. Accordingly, PgA
can directly conduct a reliable and efficient attack on deep hashing-based
retrieval by maximizing the similarity between the hash code of the adversarial
example and the pharos code. Extensive experiments on the benchmark datasets
verify that the proposed algorithm outperforms the prior state-of-the-arts in
both attack strength and speed.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 15:36:19 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Wang",
"Xunguang",
""
],
[
"Bai",
"Jiawang",
""
],
[
"Xu",
"Xinyue",
""
],
[
"Li",
"Xiaomeng",
""
]
] | TITLE: Reliable and Efficient Evaluation of Adversarial Robustness for Deep
Hashing-Based Retrieval
ABSTRACT: Deep hashing has been extensively applied to massive image retrieval due to
its efficiency and effectiveness. Recently, several adversarial attacks have
been presented to reveal the vulnerability of deep hashing models against
adversarial examples. However, existing attack methods suffer from degraded
performance or inefficiency because they underutilize the semantic relations
between original samples or spend a lot of time learning these relations with a
deep neural network. In this paper, we propose a novel Pharos-guided Attack,
dubbed PgA, to evaluate the adversarial robustness of deep hashing networks
reliably and efficiently. Specifically, we design pharos code to represent the
semantics of the benign image, which preserves the similarity to semantically
relevant samples and dissimilarity to irrelevant ones. It is proven that we can
quickly calculate the pharos code via a simple math formula. Accordingly, PgA
can directly conduct a reliable and efficient attack on deep hashing-based
retrieval by maximizing the similarity between the hash code of the adversarial
example and the pharos code. Extensive experiments on the benchmark datasets
verify that the proposed algorithm outperforms the prior state-of-the-arts in
both attack strength and speed.
| no_new_dataset | 0.943086 |
2303.12669 | Paul Gavrikov | Paul Gavrikov, Janis Keuper, Margret Keuper | An Extended Study of Human-like Behavior under Adversarial Training | 6 pages, accepted at the CVPR 2023 Workshop "The 3rd Workshop of
Adversarial Machine Learning on Computer Vision: Art of Robustness" | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural networks have a number of shortcomings. Amongst the severest ones is
the sensitivity to distribution shifts which allows models to be easily fooled
into wrong predictions by small perturbations to inputs that are often
imperceivable to humans and do not have to carry semantic meaning. Adversarial
training poses a partial solution to address this issue by training models on
worst-case perturbations. Yet, recent work has also pointed out that the
reasoning in neural networks is different from humans. Humans identify objects
by shape, while neural nets mainly employ texture cues. Exemplarily, a model
trained on photographs will likely fail to generalize to datasets containing
sketches. Interestingly, it was also shown that adversarial training seems to
favorably increase the shift toward shape bias. In this work, we revisit this
observation and provide an extensive analysis of this effect on various
architectures, the common $\ell_2$- and $\ell_\infty$-training, and
Transformer-based models. Further, we provide a possible explanation for this
phenomenon from a frequency perspective.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 15:47:16 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Gavrikov",
"Paul",
""
],
[
"Keuper",
"Janis",
""
],
[
"Keuper",
"Margret",
""
]
] | TITLE: An Extended Study of Human-like Behavior under Adversarial Training
ABSTRACT: Neural networks have a number of shortcomings. Amongst the severest ones is
the sensitivity to distribution shifts which allows models to be easily fooled
into wrong predictions by small perturbations to inputs that are often
imperceivable to humans and do not have to carry semantic meaning. Adversarial
training poses a partial solution to address this issue by training models on
worst-case perturbations. Yet, recent work has also pointed out that the
reasoning in neural networks is different from humans. Humans identify objects
by shape, while neural nets mainly employ texture cues. Exemplarily, a model
trained on photographs will likely fail to generalize to datasets containing
sketches. Interestingly, it was also shown that adversarial training seems to
favorably increase the shift toward shape bias. In this work, we revisit this
observation and provide an extensive analysis of this effect on various
architectures, the common $\ell_2$- and $\ell_\infty$-training, and
Transformer-based models. Further, we provide a possible explanation for this
phenomenon from a frequency perspective.
| no_new_dataset | 0.948442 |
2303.12698 | Chen Zhao | Chen Zhao, Dawei Du, Anthony Hoogs, Christopher Funk | Open Set Action Recognition via Multi-Label Evidential Learning | Accepted by CVPR 2023 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing methods for open-set action recognition focus on novelty detection
that assumes video clips show a single action, which is unrealistic in the real
world. We propose a new method for open set action recognition and novelty
detection via MUlti-Label Evidential learning (MULE), that goes beyond previous
novel action detection methods by addressing the more general problems of
single or multiple actors in the same scene, with simultaneous action(s) by any
actor. Our Beta Evidential Neural Network estimates multi-action uncertainty
with Beta densities based on actor-context-object relation representations. An
evidence debiasing constraint is added to the objective function for
optimization to reduce the static bias of video representations, which can
incorrectly correlate predictions and static cues. We develop a learning
algorithm based on a primal-dual average scheme update to optimize the proposed
problem. Theoretical analysis of the optimization algorithm demonstrates the
convergence of the primal solution sequence and bounds for both the loss
function and the debiasing constraint. Uncertainty and belief-based novelty
estimation mechanisms are formulated to detect novel actions. Extensive
experiments on two real-world video datasets show that our proposed approach
achieves promising performance in single/multi-actor, single/multi-action
settings.
| [
{
"version": "v1",
"created": "Mon, 27 Feb 2023 18:34:18 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhao",
"Chen",
""
],
[
"Du",
"Dawei",
""
],
[
"Hoogs",
"Anthony",
""
],
[
"Funk",
"Christopher",
""
]
] | TITLE: Open Set Action Recognition via Multi-Label Evidential Learning
ABSTRACT: Existing methods for open-set action recognition focus on novelty detection
that assumes video clips show a single action, which is unrealistic in the real
world. We propose a new method for open set action recognition and novelty
detection via MUlti-Label Evidential learning (MULE), that goes beyond previous
novel action detection methods by addressing the more general problems of
single or multiple actors in the same scene, with simultaneous action(s) by any
actor. Our Beta Evidential Neural Network estimates multi-action uncertainty
with Beta densities based on actor-context-object relation representations. An
evidence debiasing constraint is added to the objective function for
optimization to reduce the static bias of video representations, which can
incorrectly correlate predictions and static cues. We develop a learning
algorithm based on a primal-dual average scheme update to optimize the proposed
problem. Theoretical analysis of the optimization algorithm demonstrates the
convergence of the primal solution sequence and bounds for both the loss
function and the debiasing constraint. Uncertainty and belief-based novelty
estimation mechanisms are formulated to detect novel actions. Extensive
experiments on two real-world video datasets show that our proposed approach
achieves promising performance in single/multi-actor, single/multi-action
settings.
| no_new_dataset | 0.945197 |
2303.12703 | Chao Ma | Matthew Ashman, Chao Ma, Agrin Hilmkil, Joel Jennings, Cheng Zhang | Causal Reasoning in the Presence of Latent Confounders via Neural ADMG
Learning | Camera ready version for ICLR 2023 | null | null | null | cs.LG stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Latent confounding has been a long-standing obstacle for causal reasoning
from observational data. One popular approach is to model the data using
acyclic directed mixed graphs (ADMGs), which describe ancestral relations
between variables using directed and bidirected edges. However, existing
methods using ADMGs are based on either linear functional assumptions or a
discrete search that is complicated to use and lacks computational tractability
for large datasets. In this work, we further extend the existing body of work
and develop a novel gradient-based approach to learning an ADMG with non-linear
functional relations from observational data. We first show that the presence
of latent confounding is identifiable under the assumptions of bow-free ADMGs
with non-linear additive noise models. With this insight, we propose a novel
neural causal model based on autoregressive flows for ADMG learning. This not
only enables us to determine complex causal structural relationships behind the
data in the presence of latent confounding, but also estimate their functional
relationships (hence treatment effects) simultaneously. We further validate our
approach via experiments on both synthetic and real-world datasets, and
demonstrate the competitive performance against relevant baselines.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 16:45:54 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ashman",
"Matthew",
""
],
[
"Ma",
"Chao",
""
],
[
"Hilmkil",
"Agrin",
""
],
[
"Jennings",
"Joel",
""
],
[
"Zhang",
"Cheng",
""
]
] | TITLE: Causal Reasoning in the Presence of Latent Confounders via Neural ADMG
Learning
ABSTRACT: Latent confounding has been a long-standing obstacle for causal reasoning
from observational data. One popular approach is to model the data using
acyclic directed mixed graphs (ADMGs), which describe ancestral relations
between variables using directed and bidirected edges. However, existing
methods using ADMGs are based on either linear functional assumptions or a
discrete search that is complicated to use and lacks computational tractability
for large datasets. In this work, we further extend the existing body of work
and develop a novel gradient-based approach to learning an ADMG with non-linear
functional relations from observational data. We first show that the presence
of latent confounding is identifiable under the assumptions of bow-free ADMGs
with non-linear additive noise models. With this insight, we propose a novel
neural causal model based on autoregressive flows for ADMG learning. This not
only enables us to determine complex causal structural relationships behind the
data in the presence of latent confounding, but also estimate their functional
relationships (hence treatment effects) simultaneously. We further validate our
approach via experiments on both synthetic and real-world datasets, and
demonstrate the competitive performance against relevant baselines.
| no_new_dataset | 0.944944 |
2303.12707 | Kinyua Gikunda Dr. | Godfrin Ismail, Kenneth Chesoli, Golda Moni, Kinyua Gikunda | Comparison of Probabilistic Deep Learning Methods for Autism Detection | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Autism Spectrum Disorder (ASD) is one neuro developmental disorder that is
now widespread in the world. ASD persists throughout the life of an individual,
impacting the way they behave and communicate, resulting to notable deficits
consisting of social life retardation, repeated behavioural traits and a
restriction in their interests. Early detection of the disorder helps in the
onset treatment and helps one to lead a normal life. There are clinical
approaches used in detection of autism, relying on behavioural data and in
worst cases, neuroimaging. Quantitative methods involving machine learning have
been studied and developed to overcome issues with clinical approaches. These
quantitative methods rely on machine learning, with some complex methods based
on deep learning developed to accelerate detection and diagnosis of ASD. These
literature is aimed at exploring most state-of-the-art probabilistic methods in
use today, characterizing them with the type of dataset they're most applied
on, their accuracy according to their novel research and how well they are
suited in ASD classification. The findings will purposely serve as a benchmark
in selection of the model to use when performing ASD detection.
| [
{
"version": "v1",
"created": "Thu, 9 Mar 2023 17:49:37 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Ismail",
"Godfrin",
""
],
[
"Chesoli",
"Kenneth",
""
],
[
"Moni",
"Golda",
""
],
[
"Gikunda",
"Kinyua",
""
]
] | TITLE: Comparison of Probabilistic Deep Learning Methods for Autism Detection
ABSTRACT: Autism Spectrum Disorder (ASD) is one neuro developmental disorder that is
now widespread in the world. ASD persists throughout the life of an individual,
impacting the way they behave and communicate, resulting to notable deficits
consisting of social life retardation, repeated behavioural traits and a
restriction in their interests. Early detection of the disorder helps in the
onset treatment and helps one to lead a normal life. There are clinical
approaches used in detection of autism, relying on behavioural data and in
worst cases, neuroimaging. Quantitative methods involving machine learning have
been studied and developed to overcome issues with clinical approaches. These
quantitative methods rely on machine learning, with some complex methods based
on deep learning developed to accelerate detection and diagnosis of ASD. These
literature is aimed at exploring most state-of-the-art probabilistic methods in
use today, characterizing them with the type of dataset they're most applied
on, their accuracy according to their novel research and how well they are
suited in ASD classification. The findings will purposely serve as a benchmark
in selection of the model to use when performing ASD detection.
| no_new_dataset | 0.938407 |
2303.12719 | Jurdana Masuma Iqrah | Jurdana Masuma Iqrah, Younghyun Koo, Wei Wang, Hongjie Xie and Sushil
Prasad | Toward Polar Sea-Ice Classification using Color-based Segmentation and
Auto-labeling of Sentinel-2 Imagery to Train an Efficient Deep Learning Model | 2nd Annual AAAI Workshop on AI to Accelerate Science and Engineering
(AI2ASE), February 2023 | null | null | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Global warming is an urgent issue that is generating catastrophic
environmental changes, such as the melting of sea ice and glaciers,
particularly in the polar regions. The melting pattern and retreat of polar sea
ice cover is an essential indicator of global warming. The Sentinel-2 satellite
(S2) captures high-resolution optical imagery over the polar regions. This
research aims at developing a robust and effective system for classifying polar
sea ice as thick or snow-covered, young or thin, or open water using S2 images.
A key challenge is the lack of labeled S2 training data to serve as the ground
truth. We demonstrate a method with high precision to segment and automatically
label the S2 images based on suitably determined color thresholds and employ
these auto-labeled data to train a U-Net machine model (a fully convolutional
neural network), yielding good classification accuracy. Evaluation results over
S2 data from the polar summer season in the Ross Sea region of the Antarctic
show that the U-Net model trained on auto-labeled data has an accuracy of
90.18% over the original S2 images, whereas the U-Net model trained on manually
labeled data has an accuracy of 91.39%. Filtering out the thin clouds and
shadows from the S2 images further improves U-Net's accuracy, respectively, to
98.97% for auto-labeled and 98.40% for manually labeled training datasets.
| [
{
"version": "v1",
"created": "Wed, 8 Mar 2023 19:09:22 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Iqrah",
"Jurdana Masuma",
""
],
[
"Koo",
"Younghyun",
""
],
[
"Wang",
"Wei",
""
],
[
"Xie",
"Hongjie",
""
],
[
"Prasad",
"Sushil",
""
]
] | TITLE: Toward Polar Sea-Ice Classification using Color-based Segmentation and
Auto-labeling of Sentinel-2 Imagery to Train an Efficient Deep Learning Model
ABSTRACT: Global warming is an urgent issue that is generating catastrophic
environmental changes, such as the melting of sea ice and glaciers,
particularly in the polar regions. The melting pattern and retreat of polar sea
ice cover is an essential indicator of global warming. The Sentinel-2 satellite
(S2) captures high-resolution optical imagery over the polar regions. This
research aims at developing a robust and effective system for classifying polar
sea ice as thick or snow-covered, young or thin, or open water using S2 images.
A key challenge is the lack of labeled S2 training data to serve as the ground
truth. We demonstrate a method with high precision to segment and automatically
label the S2 images based on suitably determined color thresholds and employ
these auto-labeled data to train a U-Net machine model (a fully convolutional
neural network), yielding good classification accuracy. Evaluation results over
S2 data from the polar summer season in the Ross Sea region of the Antarctic
show that the U-Net model trained on auto-labeled data has an accuracy of
90.18% over the original S2 images, whereas the U-Net model trained on manually
labeled data has an accuracy of 91.39%. Filtering out the thin clouds and
shadows from the S2 images further improves U-Net's accuracy, respectively, to
98.97% for auto-labeled and 98.40% for manually labeled training datasets.
| no_new_dataset | 0.954732 |
2303.12724 | Yu Zhang | Yulong Zhang, Shuhao Chen, Yu Zhang, Jiangang Lu | Diffusion-based Target Sampler for Unsupervised Domain Adaptation | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Limited transferability hinders the performance of deep learning models when
applied to new application scenarios. Recently, unsupervised domain adaptation
(UDA) has achieved significant progress in addressing this issue via learning
domain-invariant features. However, large domain shifts and the sample scarcity
in the target domain make existing UDA methods achieve suboptimal performance.
To alleviate these issues, we propose a plug-and-play Diffusion-based Target
Sampler (DTS) to generate high fidelity and diversity pseudo target samples. By
introducing class-conditional information, the labels of the generated target
samples can be controlled. The generated samples can well simulate the data
distribution of the target domain and help existing UDA methods transfer from
the source domain to the target domain more easily, thus improving the transfer
performance. Extensive experiments on various benchmarks demonstrate that the
performance of existing UDA methods can be greatly improved through the
proposed DTS method.
| [
{
"version": "v1",
"created": "Fri, 17 Mar 2023 02:07:43 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Zhang",
"Yulong",
""
],
[
"Chen",
"Shuhao",
""
],
[
"Zhang",
"Yu",
""
],
[
"Lu",
"Jiangang",
""
]
] | TITLE: Diffusion-based Target Sampler for Unsupervised Domain Adaptation
ABSTRACT: Limited transferability hinders the performance of deep learning models when
applied to new application scenarios. Recently, unsupervised domain adaptation
(UDA) has achieved significant progress in addressing this issue via learning
domain-invariant features. However, large domain shifts and the sample scarcity
in the target domain make existing UDA methods achieve suboptimal performance.
To alleviate these issues, we propose a plug-and-play Diffusion-based Target
Sampler (DTS) to generate high fidelity and diversity pseudo target samples. By
introducing class-conditional information, the labels of the generated target
samples can be controlled. The generated samples can well simulate the data
distribution of the target domain and help existing UDA methods transfer from
the source domain to the target domain more easily, thus improving the transfer
performance. Extensive experiments on various benchmarks demonstrate that the
performance of existing UDA methods can be greatly improved through the
proposed DTS method.
| no_new_dataset | 0.941761 |
2303.12725 | Zhipeng Chang | Zhipeng Chang, Ruiling Ma, Wenliang Jia | Pedestrain detection for low-light vision proposal | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | The demand for pedestrian detection has created a challenging problem for
various visual tasks such as image fusion. As infrared images can capture
thermal radiation information, image fusion between infrared and visible images
could significantly improve target detection under environmental limitations.
In our project, we would approach by preprocessing our dataset with image
fusion technique, then using Vision Transformer model to detect pedestrians
from the fused images. During the evaluation procedure, a comparison would be
made between YOLOv5 and the revised ViT model performance on our fused images
| [
{
"version": "v1",
"created": "Fri, 17 Mar 2023 04:13:58 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Chang",
"Zhipeng",
""
],
[
"Ma",
"Ruiling",
""
],
[
"Jia",
"Wenliang",
""
]
] | TITLE: Pedestrain detection for low-light vision proposal
ABSTRACT: The demand for pedestrian detection has created a challenging problem for
various visual tasks such as image fusion. As infrared images can capture
thermal radiation information, image fusion between infrared and visible images
could significantly improve target detection under environmental limitations.
In our project, we would approach by preprocessing our dataset with image
fusion technique, then using Vision Transformer model to detect pedestrians
from the fused images. During the evaluation procedure, a comparison would be
made between YOLOv5 and the revised ViT model performance on our fused images
| no_new_dataset | 0.941547 |
2303.12728 | Akshansh Gupta | Somsukla Maiti and Akshansh Gupta | LocalEyenet: Deep Attention framework for Localization of Eyes | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Development of human machine interface has become a necessity for modern day
machines to catalyze more autonomy and more efficiency. Gaze driven human
intervention is an effective and convenient option for creating an interface to
alleviate human errors. Facial landmark detection is very crucial for designing
a robust gaze detection system. Regression based methods capacitate good
spatial localization of the landmarks corresponding to different parts of the
faces. But there are still scope of improvements which have been addressed by
incorporating attention.
In this paper, we have proposed a deep coarse-to-fine architecture called
LocalEyenet for localization of only the eye regions that can be trained
end-to-end. The model architecture, build on stacked hourglass backbone, learns
the self-attention in feature maps which aids in preserving global as well as
local spatial dependencies in face image. We have incorporated deep layer
aggregation in each hourglass to minimize the loss of attention over the depth
of architecture. Our model shows good generalization ability in cross-dataset
evaluation and in real-time localization of eyes.
| [
{
"version": "v1",
"created": "Mon, 13 Mar 2023 06:35:45 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Maiti",
"Somsukla",
""
],
[
"Gupta",
"Akshansh",
""
]
] | TITLE: LocalEyenet: Deep Attention framework for Localization of Eyes
ABSTRACT: Development of human machine interface has become a necessity for modern day
machines to catalyze more autonomy and more efficiency. Gaze driven human
intervention is an effective and convenient option for creating an interface to
alleviate human errors. Facial landmark detection is very crucial for designing
a robust gaze detection system. Regression based methods capacitate good
spatial localization of the landmarks corresponding to different parts of the
faces. But there are still scope of improvements which have been addressed by
incorporating attention.
In this paper, we have proposed a deep coarse-to-fine architecture called
LocalEyenet for localization of only the eye regions that can be trained
end-to-end. The model architecture, build on stacked hourglass backbone, learns
the self-attention in feature maps which aids in preserving global as well as
local spatial dependencies in face image. We have incorporated deep layer
aggregation in each hourglass to minimize the loss of attention over the depth
of architecture. Our model shows good generalization ability in cross-dataset
evaluation and in real-time localization of eyes.
| no_new_dataset | 0.948917 |
2303.12730 | Joshua Power | Joshua Power, Derek Jacoby, Marc-Antoine Drouin, Guillaume Durand,
Yvonne Coady, Julian Meng | Toward Data-Driven Glare Classification and Prediction for Marine
Megafauna Survey | 15 pages, 4 figures, 5th ICPR Workshop on Computer Vison for
Automated Analysis of Underwater Imagery (CVAUI 2022) | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Critically endangered species in Canadian North Atlantic waters are
systematically surveyed to estimate species populations which influence
governing policies. Due to its impact on policy, population accuracy is
important. This paper lays the foundation towards a data-driven glare modelling
system, which will allow surveyors to preemptively minimize glare. Surveyors
use a detection function to estimate megafauna populations which are not
explicitly seen. A goal of the research is to maximize useful imagery
collected, to that end we will use our glare model to predict glare and
optimize for glare-free data collection. To build this model, we leverage a
small labelled dataset to perform semi-supervised learning. The large dataset
is labelled with a Cascading Random Forest Model using a na\"ive
pseudo-labelling approach. A reflectance model is used, which pinpoints
features of interest, to populate our datasets which allows for context-aware
machine learning models. The pseudo-labelled dataset is used on two models: a
Multilayer Perceptron and a Recurrent Neural Network. With this paper, we lay
the foundation for data-driven mission planning; a glare modelling system which
allows surveyors to preemptively minimize glare and reduces survey reliance on
the detection function as an estimator of whale populations during periods of
poor subsurface visibility.
| [
{
"version": "v1",
"created": "Fri, 3 Mar 2023 18:46:19 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Power",
"Joshua",
""
],
[
"Jacoby",
"Derek",
""
],
[
"Drouin",
"Marc-Antoine",
""
],
[
"Durand",
"Guillaume",
""
],
[
"Coady",
"Yvonne",
""
],
[
"Meng",
"Julian",
""
]
] | TITLE: Toward Data-Driven Glare Classification and Prediction for Marine
Megafauna Survey
ABSTRACT: Critically endangered species in Canadian North Atlantic waters are
systematically surveyed to estimate species populations which influence
governing policies. Due to its impact on policy, population accuracy is
important. This paper lays the foundation towards a data-driven glare modelling
system, which will allow surveyors to preemptively minimize glare. Surveyors
use a detection function to estimate megafauna populations which are not
explicitly seen. A goal of the research is to maximize useful imagery
collected, to that end we will use our glare model to predict glare and
optimize for glare-free data collection. To build this model, we leverage a
small labelled dataset to perform semi-supervised learning. The large dataset
is labelled with a Cascading Random Forest Model using a na\"ive
pseudo-labelling approach. A reflectance model is used, which pinpoints
features of interest, to populate our datasets which allows for context-aware
machine learning models. The pseudo-labelled dataset is used on two models: a
Multilayer Perceptron and a Recurrent Neural Network. With this paper, we lay
the foundation for data-driven mission planning; a glare modelling system which
allows surveyors to preemptively minimize glare and reduces survey reliance on
the detection function as an estimator of whale populations during periods of
poor subsurface visibility.
| no_new_dataset | 0.94699 |
2303.12733 | Ryan Webster | Ryan Webster and Julien Rabin and Loic Simon and Frederic Jurie | On the De-duplication of LAION-2B | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Generative models, such as DALL-E, Midjourney, and Stable Diffusion, have
societal implications that extend beyond the field of computer science. These
models require large image databases like LAION-2B, which contain two billion
images. At this scale, manual inspection is difficult and automated analysis is
challenging. In addition, recent studies show that duplicated images pose
copyright problems for models trained on LAION2B, which hinders its usability.
This paper proposes an algorithmic chain that runs with modest compute, that
compresses CLIP features to enable efficient duplicate detection, even for vast
image volumes. Our approach demonstrates that roughly 700 million images, or
about 30\%, of LAION-2B's images are likely duplicated. Our method also
provides the histograms of duplication on this dataset, which we use to reveal
more examples of verbatim copies by Stable Diffusion and further justify the
approach. The current version of the de-duplicated set will be distributed
online.
| [
{
"version": "v1",
"created": "Fri, 17 Mar 2023 17:39:06 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Webster",
"Ryan",
""
],
[
"Rabin",
"Julien",
""
],
[
"Simon",
"Loic",
""
],
[
"Jurie",
"Frederic",
""
]
] | TITLE: On the De-duplication of LAION-2B
ABSTRACT: Generative models, such as DALL-E, Midjourney, and Stable Diffusion, have
societal implications that extend beyond the field of computer science. These
models require large image databases like LAION-2B, which contain two billion
images. At this scale, manual inspection is difficult and automated analysis is
challenging. In addition, recent studies show that duplicated images pose
copyright problems for models trained on LAION2B, which hinders its usability.
This paper proposes an algorithmic chain that runs with modest compute, that
compresses CLIP features to enable efficient duplicate detection, even for vast
image volumes. Our approach demonstrates that roughly 700 million images, or
about 30\%, of LAION-2B's images are likely duplicated. Our method also
provides the histograms of duplication on this dataset, which we use to reveal
more examples of verbatim copies by Stable Diffusion and further justify the
approach. The current version of the de-duplicated set will be distributed
online.
| no_new_dataset | 0.945096 |
2303.12734 | Sepehr Janghorbani | Sepehr Janghorbani and Gerard de Melo | MultiModal Bias: Introducing a Framework for Stereotypical Bias
Assessment beyond Gender and Race in Vision Language Models | null | null | null | null | cs.CV cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent breakthroughs in self supervised training have led to a new class of
pretrained vision language models. While there have been investigations of bias
in multimodal models, they have mostly focused on gender and racial bias,
giving much less attention to other relevant groups, such as minorities with
regard to religion, nationality, sexual orientation, or disabilities. This is
mainly due to lack of suitable benchmarks for such groups. We seek to address
this gap by providing a visual and textual bias benchmark called MMBias,
consisting of around 3,800 images and phrases covering 14 population subgroups.
We utilize this dataset to assess bias in several prominent self supervised
multimodal models, including CLIP, ALBEF, and ViLT. Our results show that these
models demonstrate meaningful bias favoring certain groups. Finally, we
introduce a debiasing method designed specifically for such large pre-trained
models that can be applied as a post-processing step to mitigate bias, while
preserving the remaining accuracy of the model.
| [
{
"version": "v1",
"created": "Thu, 16 Mar 2023 17:36:37 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Janghorbani",
"Sepehr",
""
],
[
"de Melo",
"Gerard",
""
]
] | TITLE: MultiModal Bias: Introducing a Framework for Stereotypical Bias
Assessment beyond Gender and Race in Vision Language Models
ABSTRACT: Recent breakthroughs in self supervised training have led to a new class of
pretrained vision language models. While there have been investigations of bias
in multimodal models, they have mostly focused on gender and racial bias,
giving much less attention to other relevant groups, such as minorities with
regard to religion, nationality, sexual orientation, or disabilities. This is
mainly due to lack of suitable benchmarks for such groups. We seek to address
this gap by providing a visual and textual bias benchmark called MMBias,
consisting of around 3,800 images and phrases covering 14 population subgroups.
We utilize this dataset to assess bias in several prominent self supervised
multimodal models, including CLIP, ALBEF, and ViLT. Our results show that these
models demonstrate meaningful bias favoring certain groups. Finally, we
introduce a debiasing method designed specifically for such large pre-trained
models that can be applied as a post-processing step to mitigate bias, while
preserving the remaining accuracy of the model.
| new_dataset | 0.961134 |
2303.12747 | Guang Yang | Xiaodan Xing, Giorgos Papanastasiou, Simon Walsh, Guang Yang | Less is More: Unsupervised Mask-guided Annotated CT Image Synthesis with
Minimum Manual Segmentations | 12 pages, 11 figures, accepted by IEEE TMI | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | As a pragmatic data augmentation tool, data synthesis has generally returned
dividends in performance for deep learning based medical image analysis.
However, generating corresponding segmentation masks for synthetic medical
images is laborious and subjective. To obtain paired synthetic medical images
and segmentations, conditional generative models that use segmentation masks as
synthesis conditions were proposed. However, these segmentation
mask-conditioned generative models still relied on large, varied, and labeled
training datasets, and they could only provide limited constraints on human
anatomical structures, leading to unrealistic image features. Moreover, the
invariant pixel-level conditions could reduce the variety of synthetic lesions
and thus reduce the efficacy of data augmentation. To address these issues, in
this work, we propose a novel strategy for medical image synthesis, namely
Unsupervised Mask (UM)-guided synthesis, to obtain both synthetic images and
segmentations using limited manual segmentation labels. We first develop a
superpixel based algorithm to generate unsupervised structural guidance and
then design a conditional generative model to synthesize images and annotations
simultaneously from those unsupervised masks in a semi-supervised multi-task
setting. In addition, we devise a multi-scale multi-task Fr\'echet Inception
Distance (MM-FID) and multi-scale multi-task standard deviation (MM-STD) to
harness both fidelity and variety evaluations of synthetic CT images. With
multiple analyses on different scales, we could produce stable image quality
measurements with high reproducibility. Compared with the segmentation mask
guided synthesis, our UM-guided synthesis provided high-quality synthetic
images with significantly higher fidelity, variety, and utility ($p<0.05$ by
Wilcoxon Signed Ranked test).
| [
{
"version": "v1",
"created": "Sun, 19 Mar 2023 20:30:35 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Xing",
"Xiaodan",
""
],
[
"Papanastasiou",
"Giorgos",
""
],
[
"Walsh",
"Simon",
""
],
[
"Yang",
"Guang",
""
]
] | TITLE: Less is More: Unsupervised Mask-guided Annotated CT Image Synthesis with
Minimum Manual Segmentations
ABSTRACT: As a pragmatic data augmentation tool, data synthesis has generally returned
dividends in performance for deep learning based medical image analysis.
However, generating corresponding segmentation masks for synthetic medical
images is laborious and subjective. To obtain paired synthetic medical images
and segmentations, conditional generative models that use segmentation masks as
synthesis conditions were proposed. However, these segmentation
mask-conditioned generative models still relied on large, varied, and labeled
training datasets, and they could only provide limited constraints on human
anatomical structures, leading to unrealistic image features. Moreover, the
invariant pixel-level conditions could reduce the variety of synthetic lesions
and thus reduce the efficacy of data augmentation. To address these issues, in
this work, we propose a novel strategy for medical image synthesis, namely
Unsupervised Mask (UM)-guided synthesis, to obtain both synthetic images and
segmentations using limited manual segmentation labels. We first develop a
superpixel based algorithm to generate unsupervised structural guidance and
then design a conditional generative model to synthesize images and annotations
simultaneously from those unsupervised masks in a semi-supervised multi-task
setting. In addition, we devise a multi-scale multi-task Fr\'echet Inception
Distance (MM-FID) and multi-scale multi-task standard deviation (MM-STD) to
harness both fidelity and variety evaluations of synthetic CT images. With
multiple analyses on different scales, we could produce stable image quality
measurements with high reproducibility. Compared with the segmentation mask
guided synthesis, our UM-guided synthesis provided high-quality synthetic
images with significantly higher fidelity, variety, and utility ($p<0.05$ by
Wilcoxon Signed Ranked test).
| no_new_dataset | 0.956472 |
2303.12753 | Junhuan Yang | Junhuan Yang, Yi Sheng, Yuzhou Zhang, Weiwen Jiang, Lei Yang | On-Device Unsupervised Image Segmentation | To be published in Design Automation Conference (DAC) 2023 | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Along with the breakthrough of convolutional neural networks, learning-based
segmentation has emerged in many research works. Most of them are based on
supervised learning, requiring plenty of annotated data; however, to support
segmentation, a label for each pixel is required, which is obviously expensive.
As a result, the issue of lacking annotated segmentation data commonly exists.
Continuous learning is a promising way to deal with this issue; however, it
still has high demands on human labor for annotation. What's more, privacy is
highly required in segmentation data for real-world applications, which further
calls for on-device learning. In this paper, we aim to resolve the above issue
in an alternative way: Instead of supervised segmentation, we propose to
develop efficient unsupervised segmentation that can be executed on edge
devices. Based on our observation that segmentation can obtain high performance
when pixels are mapped to a high-dimension space, we for the first time bring
brain-inspired hyperdimensional computing (HDC) to the segmentation task. We
build the HDC-based unsupervised segmentation framework, namely "SegHDC". In
SegHDC, we devise a novel encoding approach that follows the Manhattan
distance. A clustering algorithm is further developed on top of the encoded
high-dimension vectors to obtain segmentation results. Experimental results
show SegHDC can significantly surpass neural network-based unsupervised
segmentation. On a standard segmentation dataset, DSB2018, SegHDC can achieve a
28.0% improvement in Intersection over Union (IoU) score; meanwhile, it
achieves over 300x speedup on Raspberry PI. What's more, for a larger size
image in the BBBC005 dataset, the existing approach cannot be accommodated to
Raspberry PI due to out of memory; on the other hand, SegHDC can obtain
segmentation results within 3 minutes while achieving a 0.9587 IoU score.
| [
{
"version": "v1",
"created": "Fri, 24 Feb 2023 00:51:17 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Yang",
"Junhuan",
""
],
[
"Sheng",
"Yi",
""
],
[
"Zhang",
"Yuzhou",
""
],
[
"Jiang",
"Weiwen",
""
],
[
"Yang",
"Lei",
""
]
] | TITLE: On-Device Unsupervised Image Segmentation
ABSTRACT: Along with the breakthrough of convolutional neural networks, learning-based
segmentation has emerged in many research works. Most of them are based on
supervised learning, requiring plenty of annotated data; however, to support
segmentation, a label for each pixel is required, which is obviously expensive.
As a result, the issue of lacking annotated segmentation data commonly exists.
Continuous learning is a promising way to deal with this issue; however, it
still has high demands on human labor for annotation. What's more, privacy is
highly required in segmentation data for real-world applications, which further
calls for on-device learning. In this paper, we aim to resolve the above issue
in an alternative way: Instead of supervised segmentation, we propose to
develop efficient unsupervised segmentation that can be executed on edge
devices. Based on our observation that segmentation can obtain high performance
when pixels are mapped to a high-dimension space, we for the first time bring
brain-inspired hyperdimensional computing (HDC) to the segmentation task. We
build the HDC-based unsupervised segmentation framework, namely "SegHDC". In
SegHDC, we devise a novel encoding approach that follows the Manhattan
distance. A clustering algorithm is further developed on top of the encoded
high-dimension vectors to obtain segmentation results. Experimental results
show SegHDC can significantly surpass neural network-based unsupervised
segmentation. On a standard segmentation dataset, DSB2018, SegHDC can achieve a
28.0% improvement in Intersection over Union (IoU) score; meanwhile, it
achieves over 300x speedup on Raspberry PI. What's more, for a larger size
image in the BBBC005 dataset, the existing approach cannot be accommodated to
Raspberry PI due to out of memory; on the other hand, SegHDC can obtain
segmentation results within 3 minutes while achieving a 0.9587 IoU score.
| no_new_dataset | 0.942718 |
2303.12756 | Chen Feng | Chen Feng, Ioannis Patras | MaskCon: Masked Contrastive Learning for Coarse-Labelled Dataset | CVPR 2023 camera-ready version. Codes are available at
https://github.com/MrChenFeng/MaskCon_CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning has achieved great success in recent years with the aid of
advanced neural network structures and large-scale human-annotated datasets.
However, it is often costly and difficult to accurately and efficiently
annotate large-scale datasets, especially for some specialized domains where
fine-grained labels are required. In this setting, coarse labels are much
easier to acquire as they do not require expert knowledge. In this work, we
propose a contrastive learning method, called $\textbf{Mask}$ed
$\textbf{Con}$trastive learning~($\textbf{MaskCon}$) to address the
under-explored problem setting, where we learn with a coarse-labelled dataset
in order to address a finer labelling problem. More specifically, within the
contrastive learning framework, for each sample our method generates
soft-labels with the aid of coarse labels against other samples and another
augmented view of the sample in question. By contrast to self-supervised
contrastive learning where only the sample's augmentations are considered hard
positives, and in supervised contrastive learning where only samples with the
same coarse labels are considered hard positives, we propose soft labels based
on sample distances, that are masked by the coarse labels. This allows us to
utilize both inter-sample relations and coarse labels. We demonstrate that our
method can obtain as special cases many existing state-of-the-art works and
that it provides tighter bounds on the generalization error. Experimentally,
our method achieves significant improvement over the current state-of-the-art
in various datasets, including CIFAR10, CIFAR100, ImageNet-1K, Standford Online
Products and Stanford Cars196 datasets. Code and annotations are available at
https://github.com/MrChenFeng/MaskCon_CVPR2023.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:08:31 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Feng",
"Chen",
""
],
[
"Patras",
"Ioannis",
""
]
] | TITLE: MaskCon: Masked Contrastive Learning for Coarse-Labelled Dataset
ABSTRACT: Deep learning has achieved great success in recent years with the aid of
advanced neural network structures and large-scale human-annotated datasets.
However, it is often costly and difficult to accurately and efficiently
annotate large-scale datasets, especially for some specialized domains where
fine-grained labels are required. In this setting, coarse labels are much
easier to acquire as they do not require expert knowledge. In this work, we
propose a contrastive learning method, called $\textbf{Mask}$ed
$\textbf{Con}$trastive learning~($\textbf{MaskCon}$) to address the
under-explored problem setting, where we learn with a coarse-labelled dataset
in order to address a finer labelling problem. More specifically, within the
contrastive learning framework, for each sample our method generates
soft-labels with the aid of coarse labels against other samples and another
augmented view of the sample in question. By contrast to self-supervised
contrastive learning where only the sample's augmentations are considered hard
positives, and in supervised contrastive learning where only samples with the
same coarse labels are considered hard positives, we propose soft labels based
on sample distances, that are masked by the coarse labels. This allows us to
utilize both inter-sample relations and coarse labels. We demonstrate that our
method can obtain as special cases many existing state-of-the-art works and
that it provides tighter bounds on the generalization error. Experimentally,
our method achieves significant improvement over the current state-of-the-art
in various datasets, including CIFAR10, CIFAR100, ImageNet-1K, Standford Online
Products and Stanford Cars196 datasets. Code and annotations are available at
https://github.com/MrChenFeng/MaskCon_CVPR2023.
| no_new_dataset | 0.947186 |
2303.12760 | Jiaming Na | Jiaming Na, Varuna De-Silva | Uncertainty Aware Active Learning for Reconfiguration of Pre-trained
Deep Object-Detection Networks for New Target Domains | null | null | null | null | cs.CV cs.HC | http://creativecommons.org/licenses/by/4.0/ | Object detection is one of the most important and fundamental aspects of
computer vision tasks, which has been broadly utilized in pose estimation,
object tracking and instance segmentation models. To obtain training data for
object detection model efficiently, many datasets opt to obtain their
unannotated data in video format and the annotator needs to draw a bounding box
around each object in the images. Annotating every frame from a video is costly
and inefficient since many frames contain very similar information for the
model to learn from. How to select the most informative frames from a video to
annotate has become a highly practical task to solve but attracted little
attention in research. In this paper, we proposed a novel active learning
algorithm for object detection models to tackle this problem. In the proposed
active learning algorithm, both classification and localization informativeness
of unlabelled data are measured and aggregated. Utilizing the temporal
information from video frames, two novel localization informativeness
measurements are proposed. Furthermore, a weight curve is proposed to avoid
querying adjacent frames. Proposed active learning algorithm with multiple
configurations was evaluated on the MuPoTS dataset and FootballPD dataset.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:14:10 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Na",
"Jiaming",
""
],
[
"De-Silva",
"Varuna",
""
]
] | TITLE: Uncertainty Aware Active Learning for Reconfiguration of Pre-trained
Deep Object-Detection Networks for New Target Domains
ABSTRACT: Object detection is one of the most important and fundamental aspects of
computer vision tasks, which has been broadly utilized in pose estimation,
object tracking and instance segmentation models. To obtain training data for
object detection model efficiently, many datasets opt to obtain their
unannotated data in video format and the annotator needs to draw a bounding box
around each object in the images. Annotating every frame from a video is costly
and inefficient since many frames contain very similar information for the
model to learn from. How to select the most informative frames from a video to
annotate has become a highly practical task to solve but attracted little
attention in research. In this paper, we proposed a novel active learning
algorithm for object detection models to tackle this problem. In the proposed
active learning algorithm, both classification and localization informativeness
of unlabelled data are measured and aggregated. Utilizing the temporal
information from video frames, two novel localization informativeness
measurements are proposed. Furthermore, a weight curve is proposed to avoid
querying adjacent frames. Proposed active learning algorithm with multiple
configurations was evaluated on the MuPoTS dataset and FootballPD dataset.
| no_new_dataset | 0.950457 |
2303.12761 | Gabriel Mittag | Gabriel Mittag, Babak Naderi, Vishak Gopal, Ross Cutler | LSTM-based Video Quality Prediction Accounting for Temporal Distortions
in Videoconferencing Calls | Accepted at ICASSP 2023 | null | null | null | eess.IV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Current state-of-the-art video quality models, such as VMAF, give excellent
prediction results by comparing the degraded video with its reference video.
However, they do not consider temporal distortions (e.g., frame freezes or
skips) that occur during videoconferencing calls. In this paper, we present a
data-driven approach for modeling such distortions automatically by training an
LSTM with subjective quality ratings labeled via crowdsourcing. The videos were
collected from live videoconferencing calls in 83 different network conditions.
We applied QR codes as markers on the source videos to create aligned
references and compute temporal features based on the alignment vectors. Using
these features together with VMAF core features, our proposed model achieves a
PCC of 0.99 on the validation set. Furthermore, our model outputs per-frame
quality that gives detailed insight into the cause of video quality
impairments. The VCM model and dataset are open-sourced at
https://github.com/microsoft/Video_Call_MOS.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:14:38 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Mittag",
"Gabriel",
""
],
[
"Naderi",
"Babak",
""
],
[
"Gopal",
"Vishak",
""
],
[
"Cutler",
"Ross",
""
]
] | TITLE: LSTM-based Video Quality Prediction Accounting for Temporal Distortions
in Videoconferencing Calls
ABSTRACT: Current state-of-the-art video quality models, such as VMAF, give excellent
prediction results by comparing the degraded video with its reference video.
However, they do not consider temporal distortions (e.g., frame freezes or
skips) that occur during videoconferencing calls. In this paper, we present a
data-driven approach for modeling such distortions automatically by training an
LSTM with subjective quality ratings labeled via crowdsourcing. The videos were
collected from live videoconferencing calls in 83 different network conditions.
We applied QR codes as markers on the source videos to create aligned
references and compute temporal features based on the alignment vectors. Using
these features together with VMAF core features, our proposed model achieves a
PCC of 0.99 on the validation set. Furthermore, our model outputs per-frame
quality that gives detailed insight into the cause of video quality
impairments. The VCM model and dataset are open-sourced at
https://github.com/microsoft/Video_Call_MOS.
| no_new_dataset | 0.946843 |
2303.12766 | Xin Lai | Xin Lai, Yukang Chen, Fanbin Lu, Jianhui Liu, Jiaya Jia | Spherical Transformer for LiDAR-based 3D Recognition | Accepted to CVPR 2023. Code is available at
https://github.com/dvlab-research/SphereFormer.git | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | LiDAR-based 3D point cloud recognition has benefited various applications.
Without specially considering the LiDAR point distribution, most current
methods suffer from information disconnection and limited receptive field,
especially for the sparse distant points. In this work, we study the
varying-sparsity distribution of LiDAR points and present SphereFormer to
directly aggregate information from dense close points to the sparse distant
ones. We design radial window self-attention that partitions the space into
multiple non-overlapping narrow and long windows. It overcomes the
disconnection issue and enlarges the receptive field smoothly and dramatically,
which significantly boosts the performance of sparse distant points. Moreover,
to fit the narrow and long windows, we propose exponential splitting to yield
fine-grained position encoding and dynamic feature selection to increase model
representation ability. Notably, our method ranks 1st on both nuScenes and
SemanticKITTI semantic segmentation benchmarks with 81.9% and 74.8% mIoU,
respectively. Also, we achieve the 3rd place on nuScenes object detection
benchmark with 72.8% NDS and 68.5% mAP. Code is available at
https://github.com/dvlab-research/SphereFormer.git.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:30:14 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Lai",
"Xin",
""
],
[
"Chen",
"Yukang",
""
],
[
"Lu",
"Fanbin",
""
],
[
"Liu",
"Jianhui",
""
],
[
"Jia",
"Jiaya",
""
]
] | TITLE: Spherical Transformer for LiDAR-based 3D Recognition
ABSTRACT: LiDAR-based 3D point cloud recognition has benefited various applications.
Without specially considering the LiDAR point distribution, most current
methods suffer from information disconnection and limited receptive field,
especially for the sparse distant points. In this work, we study the
varying-sparsity distribution of LiDAR points and present SphereFormer to
directly aggregate information from dense close points to the sparse distant
ones. We design radial window self-attention that partitions the space into
multiple non-overlapping narrow and long windows. It overcomes the
disconnection issue and enlarges the receptive field smoothly and dramatically,
which significantly boosts the performance of sparse distant points. Moreover,
to fit the narrow and long windows, we propose exponential splitting to yield
fine-grained position encoding and dynamic feature selection to increase model
representation ability. Notably, our method ranks 1st on both nuScenes and
SemanticKITTI semantic segmentation benchmarks with 81.9% and 74.8% mIoU,
respectively. Also, we achieve the 3rd place on nuScenes object detection
benchmark with 72.8% NDS and 68.5% mAP. Code is available at
https://github.com/dvlab-research/SphereFormer.git.
| no_new_dataset | 0.951953 |
2303.12772 | Tasnim Sakib Apon | Ramisa Anan, Tasnim Sakib Apon, Zeba Tahsin Hossain, Elizabeth Antora
Modhu, Sudipta Mondal, MD. Golam Rabiul Alam | Interpretable Bangla Sarcasm Detection using BERT and Explainable AI | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A positive phrase or a sentence with an underlying negative motive is usually
defined as sarcasm that is widely used in today's social media platforms such
as Facebook, Twitter, Reddit, etc. In recent times active users in social media
platforms are increasing dramatically which raises the need for an automated
NLP-based system that can be utilized in various tasks such as determining
market demand, sentiment analysis, threat detection, etc. However, since
sarcasm usually implies the opposite meaning and its detection is frequently a
challenging issue, data meaning extraction through an NLP-based model becomes
more complicated. As a result, there has been a lot of study on sarcasm
detection in English over the past several years, and there's been a noticeable
improvement and yet sarcasm detection in the Bangla language's state remains
the same. In this article, we present a BERT-based system that can achieve
99.60\% while the utilized traditional machine learning algorithms are only
capable of achieving 89.93\%. Additionally, we have employed Local
Interpretable Model-Agnostic Explanations that introduce explainability to our
system. Moreover, we have utilized a newly collected bangla sarcasm dataset,
BanglaSarc that was constructed specifically for the evaluation of this study.
This dataset consists of fresh records of sarcastic and non-sarcastic comments,
the majority of which are acquired from Facebook and YouTube comment sections.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:35:35 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Anan",
"Ramisa",
""
],
[
"Apon",
"Tasnim Sakib",
""
],
[
"Hossain",
"Zeba Tahsin",
""
],
[
"Modhu",
"Elizabeth Antora",
""
],
[
"Mondal",
"Sudipta",
""
],
[
"Alam",
"MD. Golam Rabiul",
""
]
] | TITLE: Interpretable Bangla Sarcasm Detection using BERT and Explainable AI
ABSTRACT: A positive phrase or a sentence with an underlying negative motive is usually
defined as sarcasm that is widely used in today's social media platforms such
as Facebook, Twitter, Reddit, etc. In recent times active users in social media
platforms are increasing dramatically which raises the need for an automated
NLP-based system that can be utilized in various tasks such as determining
market demand, sentiment analysis, threat detection, etc. However, since
sarcasm usually implies the opposite meaning and its detection is frequently a
challenging issue, data meaning extraction through an NLP-based model becomes
more complicated. As a result, there has been a lot of study on sarcasm
detection in English over the past several years, and there's been a noticeable
improvement and yet sarcasm detection in the Bangla language's state remains
the same. In this article, we present a BERT-based system that can achieve
99.60\% while the utilized traditional machine learning algorithms are only
capable of achieving 89.93\%. Additionally, we have employed Local
Interpretable Model-Agnostic Explanations that introduce explainability to our
system. Moreover, we have utilized a newly collected bangla sarcasm dataset,
BanglaSarc that was constructed specifically for the evaluation of this study.
This dataset consists of fresh records of sarcastic and non-sarcastic comments,
the majority of which are acquired from Facebook and YouTube comment sections.
| new_dataset | 0.960473 |
2303.12793 | Fangyun Wei | Yiting Cheng, Fangyun Wei, Jianmin Bao, Dong Chen, Wenqiang Zhang | CiCo: Domain-Aware Sign Language Retrieval via Cross-Lingual Contrastive
Learning | Accepted by CVPR 2023. Code and models are available at:
https://github.com/FangyunWei/SLRT | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work focuses on sign language retrieval-a recently proposed task for
sign language understanding. Sign language retrieval consists of two sub-tasks:
text-to-sign-video (T2V) retrieval and sign-video-to-text (V2T) retrieval.
Different from traditional video-text retrieval, sign language videos, not only
contain visual signals but also carry abundant semantic meanings by themselves
due to the fact that sign languages are also natural languages. Considering
this character, we formulate sign language retrieval as a cross-lingual
retrieval problem as well as a video-text retrieval task. Concretely, we take
into account the linguistic properties of both sign languages and natural
languages, and simultaneously identify the fine-grained cross-lingual (i.e.,
sign-to-word) mappings while contrasting the texts and the sign videos in a
joint embedding space. This process is termed as cross-lingual contrastive
learning. Another challenge is raised by the data scarcity issue-sign language
datasets are orders of magnitude smaller in scale than that of speech
recognition. We alleviate this issue by adopting a domain-agnostic sign encoder
pre-trained on large-scale sign videos into the target domain via
pseudo-labeling. Our framework, termed as domain-aware sign language retrieval
via Cross-lingual Contrastive learning or CiCo for short, outperforms the
pioneering method by large margins on various datasets, e.g., +22.4 T2V and
+28.0 V2T R@1 improvements on How2Sign dataset, and +13.7 T2V and +17.1 V2T R@1
improvements on PHOENIX-2014T dataset. Code and models are available at:
https://github.com/FangyunWei/SLRT.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 17:59:59 GMT"
}
] | 2023-03-23T00:00:00 | [
[
"Cheng",
"Yiting",
""
],
[
"Wei",
"Fangyun",
""
],
[
"Bao",
"Jianmin",
""
],
[
"Chen",
"Dong",
""
],
[
"Zhang",
"Wenqiang",
""
]
] | TITLE: CiCo: Domain-Aware Sign Language Retrieval via Cross-Lingual Contrastive
Learning
ABSTRACT: This work focuses on sign language retrieval-a recently proposed task for
sign language understanding. Sign language retrieval consists of two sub-tasks:
text-to-sign-video (T2V) retrieval and sign-video-to-text (V2T) retrieval.
Different from traditional video-text retrieval, sign language videos, not only
contain visual signals but also carry abundant semantic meanings by themselves
due to the fact that sign languages are also natural languages. Considering
this character, we formulate sign language retrieval as a cross-lingual
retrieval problem as well as a video-text retrieval task. Concretely, we take
into account the linguistic properties of both sign languages and natural
languages, and simultaneously identify the fine-grained cross-lingual (i.e.,
sign-to-word) mappings while contrasting the texts and the sign videos in a
joint embedding space. This process is termed as cross-lingual contrastive
learning. Another challenge is raised by the data scarcity issue-sign language
datasets are orders of magnitude smaller in scale than that of speech
recognition. We alleviate this issue by adopting a domain-agnostic sign encoder
pre-trained on large-scale sign videos into the target domain via
pseudo-labeling. Our framework, termed as domain-aware sign language retrieval
via Cross-lingual Contrastive learning or CiCo for short, outperforms the
pioneering method by large margins on various datasets, e.g., +22.4 T2V and
+28.0 V2T R@1 improvements on How2Sign dataset, and +13.7 T2V and +17.1 V2T R@1
improvements on PHOENIX-2014T dataset. Code and models are available at:
https://github.com/FangyunWei/SLRT.
| no_new_dataset | 0.949669 |
2107.02600 | Constantin Pape | Paul Hilt, Maedeh Zarvandi, Edgar Kaziakhmedov, Sourabh Bhide, Maria
Leptin, Constantin Pape, Anna Kreshuk | Stateless actor-critic for instance segmentation with high-level priors | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Instance segmentation is an important computer vision problem which remains
challenging despite impressive recent advances due to deep learning-based
methods. Given sufficient training data, fully supervised methods can yield
excellent performance, but annotation of ground-truth data remains a major
bottleneck, especially for biomedical applications where it has to be performed
by domain experts. The amount of labels required can be drastically reduced by
using rules derived from prior knowledge to guide the segmentation. However,
these rules are in general not differentiable and thus cannot be used with
existing methods. Here, we relax this requirement by using stateless actor
critic reinforcement learning, which enables non-differentiable rewards. We
formulate the instance segmentation problem as graph partitioning and the actor
critic predicts the edge weights driven by the rewards, which are based on the
conformity of segmented instances to high-level priors on object shape,
position or size. The experiments on toy and real datasets demonstrate that we
can achieve excellent performance without any direct supervision based only on
a rich set of priors.
| [
{
"version": "v1",
"created": "Tue, 6 Jul 2021 13:20:14 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 15:11:44 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Hilt",
"Paul",
""
],
[
"Zarvandi",
"Maedeh",
""
],
[
"Kaziakhmedov",
"Edgar",
""
],
[
"Bhide",
"Sourabh",
""
],
[
"Leptin",
"Maria",
""
],
[
"Pape",
"Constantin",
""
],
[
"Kreshuk",
"Anna",
""
]
] | TITLE: Stateless actor-critic for instance segmentation with high-level priors
ABSTRACT: Instance segmentation is an important computer vision problem which remains
challenging despite impressive recent advances due to deep learning-based
methods. Given sufficient training data, fully supervised methods can yield
excellent performance, but annotation of ground-truth data remains a major
bottleneck, especially for biomedical applications where it has to be performed
by domain experts. The amount of labels required can be drastically reduced by
using rules derived from prior knowledge to guide the segmentation. However,
these rules are in general not differentiable and thus cannot be used with
existing methods. Here, we relax this requirement by using stateless actor
critic reinforcement learning, which enables non-differentiable rewards. We
formulate the instance segmentation problem as graph partitioning and the actor
critic predicts the edge weights driven by the rewards, which are based on the
conformity of segmented instances to high-level priors on object shape,
position or size. The experiments on toy and real datasets demonstrate that we
can achieve excellent performance without any direct supervision based only on
a rich set of priors.
| no_new_dataset | 0.944125 |
2111.08913 | Lie Ju | Lie Ju, Zhen Yu, Lin Wang, Xin Zhao, Xin Wang, Paul Bonnington,
Zongyuan Ge | Hierarchical Knowledge Guided Learning for Real-world Retinal Diseases
Recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the real world, medical datasets often exhibit a long-tailed data
distribution (i.e., a few classes occupy the majority of the data, while most
classes have only a limited number of samples), which results in a challenging
long-tailed learning scenario. Some recently published datasets in
ophthalmology AI consist of more than 40 kinds of retinal diseases with complex
abnormalities and variable morbidity. Nevertheless, more than 30 conditions are
rarely seen in global patient cohorts. From a modeling perspective, most deep
learning models trained on these datasets may lack the ability to generalize to
rare diseases where only a few available samples are presented for training. In
addition, there may be more than one disease for the presence of the retina,
resulting in a challenging label co-occurrence scenario, also known as
\textit{multi-label}, which can cause problems when some re-sampling strategies
are applied during training. To address the above two major challenges, this
paper presents a novel method that enables the deep neural network to learn
from a long-tailed fundus database for various retinal disease recognition.
Firstly, we exploit the prior knowledge in ophthalmology to improve the feature
representation using a hierarchy-aware pre-training. Secondly, we adopt an
instance-wise class-balanced sampling strategy to address the label
co-occurrence issue under the long-tailed medical dataset scenario. Thirdly, we
introduce a novel hybrid knowledge distillation to train a less biased
representation and classifier. We conducted extensive experiments on four
databases, including two public datasets and two in-house databases with more
than one million fundus images. The experimental results demonstrate the
superiority of our proposed methods with recognition accuracy outperforming the
state-of-the-art competitors, especially for these rare diseases.
| [
{
"version": "v1",
"created": "Wed, 17 Nov 2021 05:44:39 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 06:07:43 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Ju",
"Lie",
""
],
[
"Yu",
"Zhen",
""
],
[
"Wang",
"Lin",
""
],
[
"Zhao",
"Xin",
""
],
[
"Wang",
"Xin",
""
],
[
"Bonnington",
"Paul",
""
],
[
"Ge",
"Zongyuan",
""
]
] | TITLE: Hierarchical Knowledge Guided Learning for Real-world Retinal Diseases
Recognition
ABSTRACT: In the real world, medical datasets often exhibit a long-tailed data
distribution (i.e., a few classes occupy the majority of the data, while most
classes have only a limited number of samples), which results in a challenging
long-tailed learning scenario. Some recently published datasets in
ophthalmology AI consist of more than 40 kinds of retinal diseases with complex
abnormalities and variable morbidity. Nevertheless, more than 30 conditions are
rarely seen in global patient cohorts. From a modeling perspective, most deep
learning models trained on these datasets may lack the ability to generalize to
rare diseases where only a few available samples are presented for training. In
addition, there may be more than one disease for the presence of the retina,
resulting in a challenging label co-occurrence scenario, also known as
\textit{multi-label}, which can cause problems when some re-sampling strategies
are applied during training. To address the above two major challenges, this
paper presents a novel method that enables the deep neural network to learn
from a long-tailed fundus database for various retinal disease recognition.
Firstly, we exploit the prior knowledge in ophthalmology to improve the feature
representation using a hierarchy-aware pre-training. Secondly, we adopt an
instance-wise class-balanced sampling strategy to address the label
co-occurrence issue under the long-tailed medical dataset scenario. Thirdly, we
introduce a novel hybrid knowledge distillation to train a less biased
representation and classifier. We conducted extensive experiments on four
databases, including two public datasets and two in-house databases with more
than one million fundus images. The experimental results demonstrate the
superiority of our proposed methods with recognition accuracy outperforming the
state-of-the-art competitors, especially for these rare diseases.
| no_new_dataset | 0.880335 |
2111.12273 | Bohan Zhuang | Jing Liu, Jianfei Cai, Bohan Zhuang | Sharpness-aware Quantization for Deep Neural Networks | Tech report | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Network quantization is a dominant paradigm of model compression. However,
the abrupt changes in quantized weights during training often lead to severe
loss fluctuations and result in a sharp loss landscape, making the gradients
unstable and thus degrading the performance. Recently, Sharpness-Aware
Minimization (SAM) has been proposed to smooth the loss landscape and improve
the generalization performance of the models. Nevertheless, directly applying
SAM to the quantized models can lead to perturbation mismatch or diminishment
issues, resulting in suboptimal performance. In this paper, we propose a novel
method, dubbed Sharpness-Aware Quantization (SAQ), to explore the effect of SAM
in model compression, particularly quantization for the first time.
Specifically, we first provide a unified view of quantization and SAM by
treating them as introducing quantization noises and adversarial perturbations
to the model weights, respectively. According to whether the noise and
perturbation terms depend on each other, SAQ can be formulated into three
cases, which are analyzed and compared comprehensively. Furthermore, by
introducing an efficient training strategy, SAQ only incurs a little additional
training overhead compared with the default optimizer (e.g., SGD or AdamW).
Extensive experiments on both convolutional neural networks and Transformers
across various datasets (i.e., ImageNet, CIFAR-10/100, Oxford Flowers-102,
Oxford-IIIT Pets) show that SAQ improves the generalization performance of the
quantized models, yielding the SOTA results in uniform quantization. For
example, on ImageNet, SAQ outperforms AdamW by 1.2% on the Top-1 accuracy for
4-bit ViT-B/16. Our 4-bit ResNet-50 surpasses the previous SOTA method by 0.9%
on the Top-1 accuracy.
| [
{
"version": "v1",
"created": "Wed, 24 Nov 2021 05:16:41 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Mar 2022 23:28:23 GMT"
},
{
"version": "v3",
"created": "Sat, 1 Oct 2022 00:59:20 GMT"
},
{
"version": "v4",
"created": "Mon, 21 Nov 2022 04:50:42 GMT"
},
{
"version": "v5",
"created": "Tue, 21 Mar 2023 10:31:46 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Liu",
"Jing",
""
],
[
"Cai",
"Jianfei",
""
],
[
"Zhuang",
"Bohan",
""
]
] | TITLE: Sharpness-aware Quantization for Deep Neural Networks
ABSTRACT: Network quantization is a dominant paradigm of model compression. However,
the abrupt changes in quantized weights during training often lead to severe
loss fluctuations and result in a sharp loss landscape, making the gradients
unstable and thus degrading the performance. Recently, Sharpness-Aware
Minimization (SAM) has been proposed to smooth the loss landscape and improve
the generalization performance of the models. Nevertheless, directly applying
SAM to the quantized models can lead to perturbation mismatch or diminishment
issues, resulting in suboptimal performance. In this paper, we propose a novel
method, dubbed Sharpness-Aware Quantization (SAQ), to explore the effect of SAM
in model compression, particularly quantization for the first time.
Specifically, we first provide a unified view of quantization and SAM by
treating them as introducing quantization noises and adversarial perturbations
to the model weights, respectively. According to whether the noise and
perturbation terms depend on each other, SAQ can be formulated into three
cases, which are analyzed and compared comprehensively. Furthermore, by
introducing an efficient training strategy, SAQ only incurs a little additional
training overhead compared with the default optimizer (e.g., SGD or AdamW).
Extensive experiments on both convolutional neural networks and Transformers
across various datasets (i.e., ImageNet, CIFAR-10/100, Oxford Flowers-102,
Oxford-IIIT Pets) show that SAQ improves the generalization performance of the
quantized models, yielding the SOTA results in uniform quantization. For
example, on ImageNet, SAQ outperforms AdamW by 1.2% on the Top-1 accuracy for
4-bit ViT-B/16. Our 4-bit ResNet-50 surpasses the previous SOTA method by 0.9%
on the Top-1 accuracy.
| no_new_dataset | 0.947914 |
2112.02869 | Yuanjie Gu | Yuanjie Gu, Zhibo Xiao, Yinghan Guan, Haoran Dai, Cheng Liu, Liang Xue
and Shouyu Wang | Physics Driven Deep Retinex Fusion for Adaptive Infrared and Visible
Image Fusion | 20 pages, 9 figures | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Convolutional neural networks have turned into an illustrious tool for image
fusion and super-resolution. However, their excellent performance cannot work
without large fixed-paired datasets; and additionally, these high-demanded
ground truth data always cannot be obtained easily in fusion tasks. In this
study, we show that, the structures of generative networks capture a great deal
of image feature priors, and then these priors are sufficient to reconstruct
high-quality fused super-resolution result using only low-resolution inputs. By
this way, we propose a novel self-supervised dataset-free method for adaptive
infrared (IR) and visible (VIS) image super-resolution fusion named Deep
Retinex Fusion (DRF). The key idea of DRF is first generating component priors
which are disentangled from physical model using our designed generative
networks ZipperNet, LightingNet and AdjustingNet, then combining these priors
which captured by networks via adaptive fusion loss functions based on Retinex
theory, and finally reconstructing the super-resolution fusion results.
Furthermore, in order to verify the effectiveness of our reported DRF, both
qualitative and quantitative experiments via comparing with other
state-of-the-art methods are performed using different test sets. These results
prove that, comparing with large datasets trained methods, DRF which works
without any dataset achieves the best super-resolution fusion performance; and
more importantly, DRF can adaptively balance IR and VIS information and has
good noise immunity. DRF codes are open source available at
https://github.com/GuYuanjie/Deep-Retinex-fusion.
| [
{
"version": "v1",
"created": "Mon, 6 Dec 2021 08:50:47 GMT"
},
{
"version": "v2",
"created": "Sun, 8 May 2022 11:00:53 GMT"
},
{
"version": "v3",
"created": "Sun, 29 May 2022 06:31:15 GMT"
},
{
"version": "v4",
"created": "Tue, 21 Mar 2023 06:34:09 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Gu",
"Yuanjie",
""
],
[
"Xiao",
"Zhibo",
""
],
[
"Guan",
"Yinghan",
""
],
[
"Dai",
"Haoran",
""
],
[
"Liu",
"Cheng",
""
],
[
"Xue",
"Liang",
""
],
[
"Wang",
"Shouyu",
""
]
] | TITLE: Physics Driven Deep Retinex Fusion for Adaptive Infrared and Visible
Image Fusion
ABSTRACT: Convolutional neural networks have turned into an illustrious tool for image
fusion and super-resolution. However, their excellent performance cannot work
without large fixed-paired datasets; and additionally, these high-demanded
ground truth data always cannot be obtained easily in fusion tasks. In this
study, we show that, the structures of generative networks capture a great deal
of image feature priors, and then these priors are sufficient to reconstruct
high-quality fused super-resolution result using only low-resolution inputs. By
this way, we propose a novel self-supervised dataset-free method for adaptive
infrared (IR) and visible (VIS) image super-resolution fusion named Deep
Retinex Fusion (DRF). The key idea of DRF is first generating component priors
which are disentangled from physical model using our designed generative
networks ZipperNet, LightingNet and AdjustingNet, then combining these priors
which captured by networks via adaptive fusion loss functions based on Retinex
theory, and finally reconstructing the super-resolution fusion results.
Furthermore, in order to verify the effectiveness of our reported DRF, both
qualitative and quantitative experiments via comparing with other
state-of-the-art methods are performed using different test sets. These results
prove that, comparing with large datasets trained methods, DRF which works
without any dataset achieves the best super-resolution fusion performance; and
more importantly, DRF can adaptively balance IR and VIS information and has
good noise immunity. DRF codes are open source available at
https://github.com/GuYuanjie/Deep-Retinex-fusion.
| no_new_dataset | 0.950411 |
2201.01661 | Muhammad Ali Farooq | Muhammad Ali Farooq, Waseem Shariff, Peter Corcoran | Evaluation of Thermal Imaging on Embedded GPU Platforms for Application
in Vehicular Assistance Systems | 14 pages, 9 tables, and 27 figures | Published in IEEE-TIV Journal in 2023 | 10.1109/TIV.2022.3158094 | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study is focused on evaluating the real-time performance of thermal
object detection for smart and safe vehicular systems by deploying the trained
networks on GPU & single-board EDGE-GPU computing platforms for onboard
automotive sensor suite testing. A novel large-scale thermal dataset comprising
of > 35,000 distinct frames is acquired, processed, and open-sourced in
challenging weather and environmental scenarios. The dataset is a recorded from
lost-cost yet effective uncooled LWIR thermal camera, mounted stand-alone and
on an electric vehicle to minimize mechanical vibrations. State-of-the-art
YOLO-V5 networks variants are trained using four different public datasets as
well newly acquired local dataset for optimal generalization of DNN by
employing SGD optimizer. The effectiveness of trained networks is validated on
extensive test data using various quantitative metrics which include precision,
recall curve, mean average precision, and frames per second. The smaller
network variant of YOLO is further optimized using TensorRT inference
accelerator to explicitly boost the frames per second rate. Optimized network
engine increases the frames per second rate by 3.5 times when testing on low
power edge devices thus achieving 11 fps on Nvidia Jetson Nano and 60 fps on
Nvidia Xavier NX development boards.
| [
{
"version": "v1",
"created": "Wed, 5 Jan 2022 15:36:25 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Farooq",
"Muhammad Ali",
""
],
[
"Shariff",
"Waseem",
""
],
[
"Corcoran",
"Peter",
""
]
] | TITLE: Evaluation of Thermal Imaging on Embedded GPU Platforms for Application
in Vehicular Assistance Systems
ABSTRACT: This study is focused on evaluating the real-time performance of thermal
object detection for smart and safe vehicular systems by deploying the trained
networks on GPU & single-board EDGE-GPU computing platforms for onboard
automotive sensor suite testing. A novel large-scale thermal dataset comprising
of > 35,000 distinct frames is acquired, processed, and open-sourced in
challenging weather and environmental scenarios. The dataset is a recorded from
lost-cost yet effective uncooled LWIR thermal camera, mounted stand-alone and
on an electric vehicle to minimize mechanical vibrations. State-of-the-art
YOLO-V5 networks variants are trained using four different public datasets as
well newly acquired local dataset for optimal generalization of DNN by
employing SGD optimizer. The effectiveness of trained networks is validated on
extensive test data using various quantitative metrics which include precision,
recall curve, mean average precision, and frames per second. The smaller
network variant of YOLO is further optimized using TensorRT inference
accelerator to explicitly boost the frames per second rate. Optimized network
engine increases the frames per second rate by 3.5 times when testing on low
power edge devices thus achieving 11 fps on Nvidia Jetson Nano and 60 fps on
Nvidia Xavier NX development boards.
| new_dataset | 0.977457 |
2202.06884 | Jules Sanchez | Jules Sanchez, Jean-Emmanuel Deschaud and Fran\c{c}ois Goulette | COLA: COarse LAbel pre-training for 3D semantic segmentation of sparse
LiDAR datasets | null | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by/4.0/ | Transfer learning is a proven technique in 2D computer vision to leverage the
large amount of data available and achieve high performance with datasets
limited in size due to the cost of acquisition or annotation. In 3D, annotation
is known to be a costly task; nevertheless, pre-training methods have only
recently been investigated. Due to this cost, unsupervised pre-training has
been heavily favored. In this work, we tackle the case of real-time 3D semantic
segmentation of sparse autonomous driving LiDAR scans. Such datasets have been
increasingly released, but each has a unique label set. We propose here an
intermediate-level label set called coarse labels, which can easily be used on
any existing and future autonomous driving datasets, thus allowing all the data
available to be leveraged at once without any additional manual labeling. This
way, we have access to a larger dataset, alongside a simple task of semantic
segmentation. With it, we introduce a new pre-training task: coarse label
pre-training, also called COLA. We thoroughly analyze the impact of COLA on
various datasets and architectures and show that it yields a noticeable
performance improvement, especially when only a small dataset is available for
the finetuning task.
| [
{
"version": "v1",
"created": "Mon, 14 Feb 2022 17:19:23 GMT"
},
{
"version": "v2",
"created": "Wed, 28 Sep 2022 12:43:34 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Mar 2023 07:12:46 GMT"
}
] | 2023-03-22T00:00:00 | [
[
"Sanchez",
"Jules",
""
],
[
"Deschaud",
"Jean-Emmanuel",
""
],
[
"Goulette",
"François",
""
]
] | TITLE: COLA: COarse LAbel pre-training for 3D semantic segmentation of sparse
LiDAR datasets
ABSTRACT: Transfer learning is a proven technique in 2D computer vision to leverage the
large amount of data available and achieve high performance with datasets
limited in size due to the cost of acquisition or annotation. In 3D, annotation
is known to be a costly task; nevertheless, pre-training methods have only
recently been investigated. Due to this cost, unsupervised pre-training has
been heavily favored. In this work, we tackle the case of real-time 3D semantic
segmentation of sparse autonomous driving LiDAR scans. Such datasets have been
increasingly released, but each has a unique label set. We propose here an
intermediate-level label set called coarse labels, which can easily be used on
any existing and future autonomous driving datasets, thus allowing all the data
available to be leveraged at once without any additional manual labeling. This
way, we have access to a larger dataset, alongside a simple task of semantic
segmentation. With it, we introduce a new pre-training task: coarse label
pre-training, also called COLA. We thoroughly analyze the impact of COLA on
various datasets and architectures and show that it yields a noticeable
performance improvement, especially when only a small dataset is available for
the finetuning task.
| no_new_dataset | 0.905239 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.