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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2303.13110 | JaeWoong Shin | Jeongun Ryu, Aaron Valero Puche, JaeWoong Shin, Seonwook Park, Biagio
Brattoli, Jinhee Lee, Wonkyung Jung, Soo Ick Cho, Kyunghyun Paeng, Chan-Young
Ock, Donggeun Yoo, S\'ergio Pereira | OCELOT: Overlapped Cell on Tissue Dataset for Histopathology | Accepted for publication at CVPR'23 | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cell detection is a fundamental task in computational pathology that can be
used for extracting high-level medical information from whole-slide images. For
accurate cell detection, pathologists often zoom out to understand the
tissue-level structures and zoom in to classify cells based on their morphology
and the surrounding context. However, there is a lack of efforts to reflect
such behaviors by pathologists in the cell detection models, mainly due to the
lack of datasets containing both cell and tissue annotations with overlapping
regions. To overcome this limitation, we propose and publicly release OCELOT, a
dataset purposely dedicated to the study of cell-tissue relationships for cell
detection in histopathology. OCELOT provides overlapping cell and tissue
annotations on images acquired from multiple organs. Within this setting, we
also propose multi-task learning approaches that benefit from learning both
cell and tissue tasks simultaneously. When compared against a model trained
only for the cell detection task, our proposed approaches improve cell
detection performance on 3 datasets: proposed OCELOT, public TIGER, and
internal CARP datasets. On the OCELOT test set in particular, we show up to
6.79 improvement in F1-score. We believe the contributions of this paper,
including the release of the OCELOT dataset at
https://lunit-io.github.io/research/publications/ocelot are a crucial starting
point toward the important research direction of incorporating cell-tissue
relationships in computation pathology.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 08:57:11 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Mar 2023 03:55:50 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ryu",
"Jeongun",
""
],
[
"Puche",
"Aaron Valero",
""
],
[
"Shin",
"JaeWoong",
""
],
[
"Park",
"Seonwook",
""
],
[
"Brattoli",
"Biagio",
""
],
[
"Lee",
"Jinhee",
""
],
[
"Jung",
"Wonkyung",
""
],
[
"Cho",
"Soo Ick",
""
],
[
"Paeng",
"Kyunghyun",
""
],
[
"Ock",
"Chan-Young",
""
],
[
"Yoo",
"Donggeun",
""
],
[
"Pereira",
"Sérgio",
""
]
] | TITLE: OCELOT: Overlapped Cell on Tissue Dataset for Histopathology
ABSTRACT: Cell detection is a fundamental task in computational pathology that can be
used for extracting high-level medical information from whole-slide images. For
accurate cell detection, pathologists often zoom out to understand the
tissue-level structures and zoom in to classify cells based on their morphology
and the surrounding context. However, there is a lack of efforts to reflect
such behaviors by pathologists in the cell detection models, mainly due to the
lack of datasets containing both cell and tissue annotations with overlapping
regions. To overcome this limitation, we propose and publicly release OCELOT, a
dataset purposely dedicated to the study of cell-tissue relationships for cell
detection in histopathology. OCELOT provides overlapping cell and tissue
annotations on images acquired from multiple organs. Within this setting, we
also propose multi-task learning approaches that benefit from learning both
cell and tissue tasks simultaneously. When compared against a model trained
only for the cell detection task, our proposed approaches improve cell
detection performance on 3 datasets: proposed OCELOT, public TIGER, and
internal CARP datasets. On the OCELOT test set in particular, we show up to
6.79 improvement in F1-score. We believe the contributions of this paper,
including the release of the OCELOT dataset at
https://lunit-io.github.io/research/publications/ocelot are a crucial starting
point toward the important research direction of incorporating cell-tissue
relationships in computation pathology.
| new_dataset | 0.968915 |
2303.13511 | Zhanghan Ke | Zhanghan Ke, Yuhao Liu, Lei Zhu, Nanxuan Zhao, Rynson W.H. Lau | Neural Preset for Color Style Transfer | Project page with demos: https://zhkkke.github.io/NeuralPreset .
Artifact-free real-time 4K color style transfer via AI-generated presets.
CVPR 2023 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we present a Neural Preset technique to address the
limitations of existing color style transfer methods, including visual
artifacts, vast memory requirement, and slow style switching speed. Our method
is based on two core designs. First, we propose Deterministic Neural Color
Mapping (DNCM) to consistently operate on each pixel via an image-adaptive
color mapping matrix, avoiding artifacts and supporting high-resolution inputs
with a small memory footprint. Second, we develop a two-stage pipeline by
dividing the task into color normalization and stylization, which allows
efficient style switching by extracting color styles as presets and reusing
them on normalized input images. Due to the unavailability of pairwise
datasets, we describe how to train Neural Preset via a self-supervised
strategy. Various advantages of Neural Preset over existing methods are
demonstrated through comprehensive evaluations. Notably, Neural Preset enables
stable 4K color style transfer in real-time without artifacts. Besides, we show
that our trained model can naturally support multiple applications without
fine-tuning, including low-light image enhancement, underwater image
correction, image dehazing, and image harmonization. Project page with demos:
https://zhkkke.github.io/NeuralPreset .
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 17:59:10 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Mar 2023 17:59:21 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ke",
"Zhanghan",
""
],
[
"Liu",
"Yuhao",
""
],
[
"Zhu",
"Lei",
""
],
[
"Zhao",
"Nanxuan",
""
],
[
"Lau",
"Rynson W. H.",
""
]
] | TITLE: Neural Preset for Color Style Transfer
ABSTRACT: In this paper, we present a Neural Preset technique to address the
limitations of existing color style transfer methods, including visual
artifacts, vast memory requirement, and slow style switching speed. Our method
is based on two core designs. First, we propose Deterministic Neural Color
Mapping (DNCM) to consistently operate on each pixel via an image-adaptive
color mapping matrix, avoiding artifacts and supporting high-resolution inputs
with a small memory footprint. Second, we develop a two-stage pipeline by
dividing the task into color normalization and stylization, which allows
efficient style switching by extracting color styles as presets and reusing
them on normalized input images. Due to the unavailability of pairwise
datasets, we describe how to train Neural Preset via a self-supervised
strategy. Various advantages of Neural Preset over existing methods are
demonstrated through comprehensive evaluations. Notably, Neural Preset enables
stable 4K color style transfer in real-time without artifacts. Besides, we show
that our trained model can naturally support multiple applications without
fine-tuning, including low-light image enhancement, underwater image
correction, image dehazing, and image harmonization. Project page with demos:
https://zhkkke.github.io/NeuralPreset .
| no_new_dataset | 0.947332 |
2303.13527 | Yuyang Wang | Yuyang Wang, Ruichen Li, Jean-R\'emy Chardonnet, Pan Hui | Dataset for predicting cybersickness from a virtual navigation task | null | null | null | null | cs.HC cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work presents a dataset collected to predict cybersickness in virtual
reality environments. The data was collected from navigation tasks in a virtual
environment designed to induce cybersickness. The dataset consists of many data
points collected from diverse participants, including physiological responses
(EDA and Heart Rate) and self-reported cybersickness symptoms. The paper will
provide a detailed description of the dataset, including the arranged
navigation task, the data collection procedures, and the data format. The
dataset will serve as a valuable resource for researchers to develop and
evaluate predictive models for cybersickness and will facilitate more research
in cybersickness mitigation.
| [
{
"version": "v1",
"created": "Tue, 7 Feb 2023 03:57:56 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wang",
"Yuyang",
""
],
[
"Li",
"Ruichen",
""
],
[
"Chardonnet",
"Jean-Rémy",
""
],
[
"Hui",
"Pan",
""
]
] | TITLE: Dataset for predicting cybersickness from a virtual navigation task
ABSTRACT: This work presents a dataset collected to predict cybersickness in virtual
reality environments. The data was collected from navigation tasks in a virtual
environment designed to induce cybersickness. The dataset consists of many data
points collected from diverse participants, including physiological responses
(EDA and Heart Rate) and self-reported cybersickness symptoms. The paper will
provide a detailed description of the dataset, including the arranged
navigation task, the data collection procedures, and the data format. The
dataset will serve as a valuable resource for researchers to develop and
evaluate predictive models for cybersickness and will facilitate more research
in cybersickness mitigation.
| new_dataset | 0.964888 |
2303.13532 | Ala-Eddine Yahiaoui | Ala-Eddine Yahiaoui, Sohaib Afifi and Hamid Afifi | Enhanced Iterated local search for the technician routing and scheduling
problem | Submitted manuscript to Computers and Operations Research journal. 34
pages, 7 figures, 6 tables | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Most public facilities in the European countries, including France, Germany,
and the UK, were built during the reconstruction projects between 1950 and
1980. Owing to the deteriorating state of such vital infrastructure has become
relatively expensive in the recent decades. A significant part of the
maintenance operation costs is spent on the technical staff. Therefore, the
optimal use of the available workforce is essential to optimize the operation
costs. This includes planning technical interventions, workload balancing,
productivity improvement, etc. In this paper, we focus on the routing of
technicians and scheduling of their tasks. We address for this purpose a
variant of the workforce scheduling problem called the technician routing and
scheduling problem (TRSP). This problem has applications in different fields,
such as transportation infrastructure (rail and road networks),
telecommunications, and sewage facilities. To solve the TRSP, we propose an
enhanced iterated local search (eILS) approach. The enhancement of the ILS
firstly includes an intensification procedure that incorporates a set of local
search operators and removal-repair heuristics crafted for the TRSP. Next, four
different mechanisms are used in the perturbation phase. Finally, an elite set
of solutions is used to extensively explore the neighborhood of local optima as
well as to enhance diversification during search space exploration. To measure
the performance of the proposed method, experiments were conducted based on
benchmark instances from the literature, and the results obtained were compared
with those of an existing method. Our method achieved very good results, since
it reached the best overall gap, which is three times lower than that of the
literature. Furthermore, eILS improved the best-known solution for $34$
instances among a total of $56$ while maintaining reasonable computational
times.
| [
{
"version": "v1",
"created": "Sun, 12 Mar 2023 23:44:49 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Yahiaoui",
"Ala-Eddine",
""
],
[
"Afifi",
"Sohaib",
""
],
[
"Afifi",
"Hamid",
""
]
] | TITLE: Enhanced Iterated local search for the technician routing and scheduling
problem
ABSTRACT: Most public facilities in the European countries, including France, Germany,
and the UK, were built during the reconstruction projects between 1950 and
1980. Owing to the deteriorating state of such vital infrastructure has become
relatively expensive in the recent decades. A significant part of the
maintenance operation costs is spent on the technical staff. Therefore, the
optimal use of the available workforce is essential to optimize the operation
costs. This includes planning technical interventions, workload balancing,
productivity improvement, etc. In this paper, we focus on the routing of
technicians and scheduling of their tasks. We address for this purpose a
variant of the workforce scheduling problem called the technician routing and
scheduling problem (TRSP). This problem has applications in different fields,
such as transportation infrastructure (rail and road networks),
telecommunications, and sewage facilities. To solve the TRSP, we propose an
enhanced iterated local search (eILS) approach. The enhancement of the ILS
firstly includes an intensification procedure that incorporates a set of local
search operators and removal-repair heuristics crafted for the TRSP. Next, four
different mechanisms are used in the perturbation phase. Finally, an elite set
of solutions is used to extensively explore the neighborhood of local optima as
well as to enhance diversification during search space exploration. To measure
the performance of the proposed method, experiments were conducted based on
benchmark instances from the literature, and the results obtained were compared
with those of an existing method. Our method achieved very good results, since
it reached the best overall gap, which is three times lower than that of the
literature. Furthermore, eILS improved the best-known solution for $34$
instances among a total of $56$ while maintaining reasonable computational
times.
| no_new_dataset | 0.947575 |
2303.13543 | Chengyu Sun | Chengyu Sun, Xing Ai, Zhihong Zhang, Edwin R Hancock | Labeled Subgraph Entropy Kernel | 9 pages,5 figures | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, kernel methods are widespread in tasks of similarity
measuring. Specifically, graph kernels are widely used in fields of
bioinformatics, chemistry and financial data analysis. However, existing
methods, especially entropy based graph kernels are subject to large
computational complexity and the negligence of node-level information. In this
paper, we propose a novel labeled subgraph entropy graph kernel, which performs
well in structural similarity assessment. We design a dynamic programming
subgraph enumeration algorithm, which effectively reduces the time complexity.
Specially, we propose labeled subgraph, which enriches substructure topology
with semantic information. Analogizing the cluster expansion process of gas
cluster in statistical mechanics, we re-derive the partition function and
calculate the global graph entropy to characterize the network. In order to
test our method, we apply several real-world datasets and assess the effects in
different tasks. To capture more experiment details, we quantitatively and
qualitatively analyze the contribution of different topology structures.
Experimental results successfully demonstrate the effectiveness of our method
which outperforms several state-of-the-art methods.
| [
{
"version": "v1",
"created": "Tue, 21 Mar 2023 12:27:21 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Sun",
"Chengyu",
""
],
[
"Ai",
"Xing",
""
],
[
"Zhang",
"Zhihong",
""
],
[
"Hancock",
"Edwin R",
""
]
] | TITLE: Labeled Subgraph Entropy Kernel
ABSTRACT: In recent years, kernel methods are widespread in tasks of similarity
measuring. Specifically, graph kernels are widely used in fields of
bioinformatics, chemistry and financial data analysis. However, existing
methods, especially entropy based graph kernels are subject to large
computational complexity and the negligence of node-level information. In this
paper, we propose a novel labeled subgraph entropy graph kernel, which performs
well in structural similarity assessment. We design a dynamic programming
subgraph enumeration algorithm, which effectively reduces the time complexity.
Specially, we propose labeled subgraph, which enriches substructure topology
with semantic information. Analogizing the cluster expansion process of gas
cluster in statistical mechanics, we re-derive the partition function and
calculate the global graph entropy to characterize the network. In order to
test our method, we apply several real-world datasets and assess the effects in
different tasks. To capture more experiment details, we quantitatively and
qualitatively analyze the contribution of different topology structures.
Experimental results successfully demonstrate the effectiveness of our method
which outperforms several state-of-the-art methods.
| no_new_dataset | 0.951278 |
2303.13547 | Aiwei Liu | Aiwei Liu, Xuming Hu, Lijie Wen, Philip S. Yu | A comprehensive evaluation of ChatGPT's zero-shot Text-to-SQL capability | 6 pages, 1 figures | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper presents the first comprehensive analysis of ChatGPT's Text-to-SQL
ability. Given the recent emergence of large-scale conversational language
model ChatGPT and its impressive capabilities in both conversational abilities
and code generation, we sought to evaluate its Text-to-SQL performance. We
conducted experiments on 12 benchmark datasets with different languages,
settings, or scenarios, and the results demonstrate that ChatGPT has strong
text-to-SQL abilities. Although there is still a gap from the current
state-of-the-art (SOTA) model performance, considering that the experiment was
conducted in a zero-shot scenario, ChatGPT's performance is still impressive.
Notably, in the ADVETA (RPL) scenario, the zero-shot ChatGPT even outperforms
the SOTA model that requires fine-tuning on the Spider dataset by 4.1\%,
demonstrating its potential for use in practical applications. To support
further research in related fields, we have made the data generated by ChatGPT
publicly available at https://github.com/THU-BPM/chatgpt-sql.
| [
{
"version": "v1",
"created": "Sun, 12 Mar 2023 04:22:01 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Liu",
"Aiwei",
""
],
[
"Hu",
"Xuming",
""
],
[
"Wen",
"Lijie",
""
],
[
"Yu",
"Philip S.",
""
]
] | TITLE: A comprehensive evaluation of ChatGPT's zero-shot Text-to-SQL capability
ABSTRACT: This paper presents the first comprehensive analysis of ChatGPT's Text-to-SQL
ability. Given the recent emergence of large-scale conversational language
model ChatGPT and its impressive capabilities in both conversational abilities
and code generation, we sought to evaluate its Text-to-SQL performance. We
conducted experiments on 12 benchmark datasets with different languages,
settings, or scenarios, and the results demonstrate that ChatGPT has strong
text-to-SQL abilities. Although there is still a gap from the current
state-of-the-art (SOTA) model performance, considering that the experiment was
conducted in a zero-shot scenario, ChatGPT's performance is still impressive.
Notably, in the ADVETA (RPL) scenario, the zero-shot ChatGPT even outperforms
the SOTA model that requires fine-tuning on the Spider dataset by 4.1\%,
demonstrating its potential for use in practical applications. To support
further research in related fields, we have made the data generated by ChatGPT
publicly available at https://github.com/THU-BPM/chatgpt-sql.
| no_new_dataset | 0.947039 |
2303.13556 | Islam Nassar | Islam Nassar, Munawar Hayat, Ehsan Abbasnejad, Hamid Rezatofighi,
Gholamreza Haffari | ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical
Consistency for Efficient Semi-supervised Learning | Accepted in CVPR2023 (highlight) | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Confidence-based pseudo-labeling is among the dominant approaches in
semi-supervised learning (SSL). It relies on including high-confidence
predictions made on unlabeled data as additional targets to train the model. We
propose ProtoCon, a novel SSL method aimed at the less-explored label-scarce
SSL where such methods usually underperform. ProtoCon refines the pseudo-labels
by leveraging their nearest neighbours' information. The neighbours are
identified as the training proceeds using an online clustering approach
operating in an embedding space trained via a prototypical loss to encourage
well-formed clusters. The online nature of ProtoCon allows it to utilise the
label history of the entire dataset in one training cycle to refine labels in
the following cycle without the need to store image embeddings. Hence, it can
seamlessly scale to larger datasets at a low cost. Finally, ProtoCon addresses
the poor training signal in the initial phase of training (due to fewer
confident predictions) by introducing an auxiliary self-supervised loss. It
delivers significant gains and faster convergence over state-of-the-art across
5 datasets, including CIFARs, ImageNet and DomainNet.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 23:51:54 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Nassar",
"Islam",
""
],
[
"Hayat",
"Munawar",
""
],
[
"Abbasnejad",
"Ehsan",
""
],
[
"Rezatofighi",
"Hamid",
""
],
[
"Haffari",
"Gholamreza",
""
]
] | TITLE: ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical
Consistency for Efficient Semi-supervised Learning
ABSTRACT: Confidence-based pseudo-labeling is among the dominant approaches in
semi-supervised learning (SSL). It relies on including high-confidence
predictions made on unlabeled data as additional targets to train the model. We
propose ProtoCon, a novel SSL method aimed at the less-explored label-scarce
SSL where such methods usually underperform. ProtoCon refines the pseudo-labels
by leveraging their nearest neighbours' information. The neighbours are
identified as the training proceeds using an online clustering approach
operating in an embedding space trained via a prototypical loss to encourage
well-formed clusters. The online nature of ProtoCon allows it to utilise the
label history of the entire dataset in one training cycle to refine labels in
the following cycle without the need to store image embeddings. Hence, it can
seamlessly scale to larger datasets at a low cost. Finally, ProtoCon addresses
the poor training signal in the initial phase of training (due to fewer
confident predictions) by introducing an auxiliary self-supervised loss. It
delivers significant gains and faster convergence over state-of-the-art across
5 datasets, including CIFARs, ImageNet and DomainNet.
| no_new_dataset | 0.949763 |
2303.13559 | Xianchao Wu | Xianchao Wu | Enhancing Unsupervised Speech Recognition with Diffusion GANs | 5 pages, 1 figure, accepted by ICASSP 2023 | null | null | null | cs.CL cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We enhance the vanilla adversarial training method for unsupervised Automatic
Speech Recognition (ASR) by a diffusion-GAN. Our model (1) injects instance
noises of various intensities to the generator's output and unlabeled reference
text which are sampled from pretrained phoneme language models with a length
constraint, (2) asks diffusion timestep-dependent discriminators to separate
them, and (3) back-propagates the gradients to update the generator.
Word/phoneme error rate comparisons with wav2vec-U under Librispeech (3.1% for
test-clean and 5.6% for test-other), TIMIT and MLS datasets, show that our
enhancement strategies work effectively.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 02:54:00 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wu",
"Xianchao",
""
]
] | TITLE: Enhancing Unsupervised Speech Recognition with Diffusion GANs
ABSTRACT: We enhance the vanilla adversarial training method for unsupervised Automatic
Speech Recognition (ASR) by a diffusion-GAN. Our model (1) injects instance
noises of various intensities to the generator's output and unlabeled reference
text which are sampled from pretrained phoneme language models with a length
constraint, (2) asks diffusion timestep-dependent discriminators to separate
them, and (3) back-propagates the gradients to update the generator.
Word/phoneme error rate comparisons with wav2vec-U under Librispeech (3.1% for
test-clean and 5.6% for test-other), TIMIT and MLS datasets, show that our
enhancement strategies work effectively.
| no_new_dataset | 0.948917 |
2303.13560 | Yue Hu | Yue Hu, Yifan Lu, Runsheng Xu, Weidi Xie, Siheng Chen, Yanfeng Wang | Collaboration Helps Camera Overtake LiDAR in 3D Detection | Accepted by CVPR23 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Camera-only 3D detection provides an economical solution with a simple
configuration for localizing objects in 3D space compared to LiDAR-based
detection systems. However, a major challenge lies in precise depth estimation
due to the lack of direct 3D measurements in the input. Many previous methods
attempt to improve depth estimation through network designs, e.g., deformable
layers and larger receptive fields. This work proposes an orthogonal direction,
improving the camera-only 3D detection by introducing multi-agent
collaborations. Our proposed collaborative camera-only 3D detection (CoCa3D)
enables agents to share complementary information with each other through
communication. Meanwhile, we optimize communication efficiency by selecting the
most informative cues. The shared messages from multiple viewpoints
disambiguate the single-agent estimated depth and complement the occluded and
long-range regions in the single-agent view. We evaluate CoCa3D in one
real-world dataset and two new simulation datasets. Results show that CoCa3D
improves previous SOTA performances by 44.21% on DAIR-V2X, 30.60% on OPV2V+,
12.59% on CoPerception-UAVs+ for AP@70. Our preliminary results show a
potential that with sufficient collaboration, the camera might overtake LiDAR
in some practical scenarios. We released the dataset and code at
https://siheng-chen.github.io/dataset/CoPerception+ and
https://github.com/MediaBrain-SJTU/CoCa3D.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 03:50:41 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Hu",
"Yue",
""
],
[
"Lu",
"Yifan",
""
],
[
"Xu",
"Runsheng",
""
],
[
"Xie",
"Weidi",
""
],
[
"Chen",
"Siheng",
""
],
[
"Wang",
"Yanfeng",
""
]
] | TITLE: Collaboration Helps Camera Overtake LiDAR in 3D Detection
ABSTRACT: Camera-only 3D detection provides an economical solution with a simple
configuration for localizing objects in 3D space compared to LiDAR-based
detection systems. However, a major challenge lies in precise depth estimation
due to the lack of direct 3D measurements in the input. Many previous methods
attempt to improve depth estimation through network designs, e.g., deformable
layers and larger receptive fields. This work proposes an orthogonal direction,
improving the camera-only 3D detection by introducing multi-agent
collaborations. Our proposed collaborative camera-only 3D detection (CoCa3D)
enables agents to share complementary information with each other through
communication. Meanwhile, we optimize communication efficiency by selecting the
most informative cues. The shared messages from multiple viewpoints
disambiguate the single-agent estimated depth and complement the occluded and
long-range regions in the single-agent view. We evaluate CoCa3D in one
real-world dataset and two new simulation datasets. Results show that CoCa3D
improves previous SOTA performances by 44.21% on DAIR-V2X, 30.60% on OPV2V+,
12.59% on CoPerception-UAVs+ for AP@70. Our preliminary results show a
potential that with sufficient collaboration, the camera might overtake LiDAR
in some practical scenarios. We released the dataset and code at
https://siheng-chen.github.io/dataset/CoPerception+ and
https://github.com/MediaBrain-SJTU/CoCa3D.
| new_dataset | 0.824885 |
2303.13561 | Yunsong Zhou | Yunsong Zhou, Quan Liu, Hongzi Zhu, Yunzhe Li, Shan Chang, and Minyi
Guo | MoGDE: Boosting Mobile Monocular 3D Object Detection with Ground Depth
Estimation | 36th Conference on Neural Information Processing Systems (NeurIPS),
2022. arXiv admin note: text overlap with arXiv:2303.13018 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Monocular 3D object detection (Mono3D) in mobile settings (e.g., on a
vehicle, a drone, or a robot) is an important yet challenging task. Due to the
near-far disparity phenomenon of monocular vision and the ever-changing camera
pose, it is hard to acquire high detection accuracy, especially for far
objects. Inspired by the insight that the depth of an object can be well
determined according to the depth of the ground where it stands, in this paper,
we propose a novel Mono3D framework, called MoGDE, which constantly estimates
the corresponding ground depth of an image and then utilizes the estimated
ground depth information to guide Mono3D. To this end, we utilize a pose
detection network to estimate the pose of the camera and then construct a
feature map portraying pixel-level ground depth according to the 3D-to-2D
perspective geometry. Moreover, to improve Mono3D with the estimated ground
depth, we design an RGB-D feature fusion network based on the transformer
structure, where the long-range self-attention mechanism is utilized to
effectively identify ground-contacting points and pin the corresponding ground
depth to the image feature map. We conduct extensive experiments on the
real-world KITTI dataset. The results demonstrate that MoGDE can effectively
improve the Mono3D accuracy and robustness for both near and far objects. MoGDE
yields the best performance compared with the state-of-the-art methods by a
large margin and is ranked number one on the KITTI 3D benchmark.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 04:06:01 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zhou",
"Yunsong",
""
],
[
"Liu",
"Quan",
""
],
[
"Zhu",
"Hongzi",
""
],
[
"Li",
"Yunzhe",
""
],
[
"Chang",
"Shan",
""
],
[
"Guo",
"Minyi",
""
]
] | TITLE: MoGDE: Boosting Mobile Monocular 3D Object Detection with Ground Depth
Estimation
ABSTRACT: Monocular 3D object detection (Mono3D) in mobile settings (e.g., on a
vehicle, a drone, or a robot) is an important yet challenging task. Due to the
near-far disparity phenomenon of monocular vision and the ever-changing camera
pose, it is hard to acquire high detection accuracy, especially for far
objects. Inspired by the insight that the depth of an object can be well
determined according to the depth of the ground where it stands, in this paper,
we propose a novel Mono3D framework, called MoGDE, which constantly estimates
the corresponding ground depth of an image and then utilizes the estimated
ground depth information to guide Mono3D. To this end, we utilize a pose
detection network to estimate the pose of the camera and then construct a
feature map portraying pixel-level ground depth according to the 3D-to-2D
perspective geometry. Moreover, to improve Mono3D with the estimated ground
depth, we design an RGB-D feature fusion network based on the transformer
structure, where the long-range self-attention mechanism is utilized to
effectively identify ground-contacting points and pin the corresponding ground
depth to the image feature map. We conduct extensive experiments on the
real-world KITTI dataset. The results demonstrate that MoGDE can effectively
improve the Mono3D accuracy and robustness for both near and far objects. MoGDE
yields the best performance compared with the state-of-the-art methods by a
large margin and is ranked number one on the KITTI 3D benchmark.
| no_new_dataset | 0.948298 |
2303.13563 | Hadjer Benmeziane | Hadjer Benmeziane, Amine Ziad Ounnoughene, Imane Hamzaoui, Younes
Bouhadjar | Skip Connections in Spiking Neural Networks: An Analysis of Their Effect
on Network Training | IPDPSW Scalable Deep Learning 2023 | null | null | null | cs.NE cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Spiking neural networks (SNNs) have gained attention as a promising
alternative to traditional artificial neural networks (ANNs) due to their
potential for energy efficiency and their ability to model spiking behavior in
biological systems. However, the training of SNNs is still a challenging
problem, and new techniques are needed to improve their performance. In this
paper, we study the impact of skip connections on SNNs and propose a
hyperparameter optimization technique that adapts models from ANN to SNN. We
demonstrate that optimizing the position, type, and number of skip connections
can significantly improve the accuracy and efficiency of SNNs by enabling
faster convergence and increasing information flow through the network. Our
results show an average +8% accuracy increase on CIFAR-10-DVS and DVS128
Gesture datasets adaptation of multiple state-of-the-art models.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 07:57:32 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Benmeziane",
"Hadjer",
""
],
[
"Ounnoughene",
"Amine Ziad",
""
],
[
"Hamzaoui",
"Imane",
""
],
[
"Bouhadjar",
"Younes",
""
]
] | TITLE: Skip Connections in Spiking Neural Networks: An Analysis of Their Effect
on Network Training
ABSTRACT: Spiking neural networks (SNNs) have gained attention as a promising
alternative to traditional artificial neural networks (ANNs) due to their
potential for energy efficiency and their ability to model spiking behavior in
biological systems. However, the training of SNNs is still a challenging
problem, and new techniques are needed to improve their performance. In this
paper, we study the impact of skip connections on SNNs and propose a
hyperparameter optimization technique that adapts models from ANN to SNN. We
demonstrate that optimizing the position, type, and number of skip connections
can significantly improve the accuracy and efficiency of SNNs by enabling
faster convergence and increasing information flow through the network. Our
results show an average +8% accuracy increase on CIFAR-10-DVS and DVS128
Gesture datasets adaptation of multiple state-of-the-art models.
| no_new_dataset | 0.952882 |
2303.13566 | Francesco Giannini | Michelangelo Diligenti, Francesco Giannini, Stefano Fioravanti,
Caterina Graziani, Moreno Falaschi, Giuseppe Marra | Enhancing Embedding Representations of Biomedical Data using Logic
Knowledge | null | null | null | null | cs.AI cs.LG q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge Graph Embeddings (KGE) have become a quite popular class of models
specifically devised to deal with ontologies and graph structure data, as they
can implicitly encode statistical dependencies between entities and relations
in a latent space. KGE techniques are particularly effective for the biomedical
domain, where it is quite common to deal with large knowledge graphs underlying
complex interactions between biological and chemical objects. Recently in the
literature, the PharmKG dataset has been proposed as one of the most
challenging knowledge graph biomedical benchmark, with hundreds of thousands of
relational facts between genes, diseases and chemicals. Despite KGEs can scale
to very large relational domains, they generally fail at representing more
complex relational dependencies between facts, like logic rules, which may be
fundamental in complex experimental settings. In this paper, we exploit logic
rules to enhance the embedding representations of KGEs on the PharmKG dataset.
To this end, we adopt Relational Reasoning Network (R2N), a recently proposed
neural-symbolic approach showing promising results on knowledge graph
completion tasks. An R2N uses the available logic rules to build a neural
architecture that reasons over KGE latent representations. In the experiments,
we show that our approach is able to significantly improve the current
state-of-the-art on the PharmKG dataset. Finally, we provide an ablation study
to experimentally compare the effect of alternative sets of rules according to
different selection criteria and varying the number of considered rules.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 13:38:21 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Diligenti",
"Michelangelo",
""
],
[
"Giannini",
"Francesco",
""
],
[
"Fioravanti",
"Stefano",
""
],
[
"Graziani",
"Caterina",
""
],
[
"Falaschi",
"Moreno",
""
],
[
"Marra",
"Giuseppe",
""
]
] | TITLE: Enhancing Embedding Representations of Biomedical Data using Logic
Knowledge
ABSTRACT: Knowledge Graph Embeddings (KGE) have become a quite popular class of models
specifically devised to deal with ontologies and graph structure data, as they
can implicitly encode statistical dependencies between entities and relations
in a latent space. KGE techniques are particularly effective for the biomedical
domain, where it is quite common to deal with large knowledge graphs underlying
complex interactions between biological and chemical objects. Recently in the
literature, the PharmKG dataset has been proposed as one of the most
challenging knowledge graph biomedical benchmark, with hundreds of thousands of
relational facts between genes, diseases and chemicals. Despite KGEs can scale
to very large relational domains, they generally fail at representing more
complex relational dependencies between facts, like logic rules, which may be
fundamental in complex experimental settings. In this paper, we exploit logic
rules to enhance the embedding representations of KGEs on the PharmKG dataset.
To this end, we adopt Relational Reasoning Network (R2N), a recently proposed
neural-symbolic approach showing promising results on knowledge graph
completion tasks. An R2N uses the available logic rules to build a neural
architecture that reasons over KGE latent representations. In the experiments,
we show that our approach is able to significantly improve the current
state-of-the-art on the PharmKG dataset. Finally, we provide an ablation study
to experimentally compare the effect of alternative sets of rules according to
different selection criteria and varying the number of considered rules.
| no_new_dataset | 0.735119 |
2303.13610 | Cheng Jiang | Todd C. Hollon, Cheng Jiang, Asadur Chowdury, Mustafa Nasir-Moin,
Akhil Kondepudi, Alexander Aabedi, Arjun Adapa, Wajd Al-Holou, Jason Heth,
Oren Sagher, Pedro Lowenstein, Maria Castro, Lisa Irina Wadiura, Georg
Widhalm, Volker Neuschmelting, David Reinecke, Niklas von Spreckelsen,
Mitchel S. Berger, Shawn L. Hervey-Jumper, John G. Golfinos, Matija Snuderl,
Sandra Camelo-Piragua, Christian Freudiger, Honglak Lee, Daniel A. Orringer | Artificial-intelligence-based molecular classification of diffuse
gliomas using rapid, label-free optical imaging | Paper published in Nature Medicine | null | 10.1038/s41591-023-02252-4 | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Molecular classification has transformed the management of brain tumors by
enabling more accurate prognostication and personalized treatment. However,
timely molecular diagnostic testing for patients with brain tumors is limited,
complicating surgical and adjuvant treatment and obstructing clinical trial
enrollment. In this study, we developed DeepGlioma, a rapid ($< 90$ seconds),
artificial-intelligence-based diagnostic screening system to streamline the
molecular diagnosis of diffuse gliomas. DeepGlioma is trained using a
multimodal dataset that includes stimulated Raman histology (SRH); a rapid,
label-free, non-consumptive, optical imaging method; and large-scale, public
genomic data. In a prospective, multicenter, international testing cohort of
patients with diffuse glioma ($n=153$) who underwent real-time SRH imaging, we
demonstrate that DeepGlioma can predict the molecular alterations used by the
World Health Organization to define the adult-type diffuse glioma taxonomy (IDH
mutation, 1p19q co-deletion and ATRX mutation), achieving a mean molecular
classification accuracy of $93.3\pm 1.6\%$. Our results represent how
artificial intelligence and optical histology can be used to provide a rapid
and scalable adjunct to wet lab methods for the molecular screening of patients
with diffuse glioma.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 18:50:18 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Hollon",
"Todd C.",
""
],
[
"Jiang",
"Cheng",
""
],
[
"Chowdury",
"Asadur",
""
],
[
"Nasir-Moin",
"Mustafa",
""
],
[
"Kondepudi",
"Akhil",
""
],
[
"Aabedi",
"Alexander",
""
],
[
"Adapa",
"Arjun",
""
],
[
"Al-Holou",
"Wajd",
""
],
[
"Heth",
"Jason",
""
],
[
"Sagher",
"Oren",
""
],
[
"Lowenstein",
"Pedro",
""
],
[
"Castro",
"Maria",
""
],
[
"Wadiura",
"Lisa Irina",
""
],
[
"Widhalm",
"Georg",
""
],
[
"Neuschmelting",
"Volker",
""
],
[
"Reinecke",
"David",
""
],
[
"von Spreckelsen",
"Niklas",
""
],
[
"Berger",
"Mitchel S.",
""
],
[
"Hervey-Jumper",
"Shawn L.",
""
],
[
"Golfinos",
"John G.",
""
],
[
"Snuderl",
"Matija",
""
],
[
"Camelo-Piragua",
"Sandra",
""
],
[
"Freudiger",
"Christian",
""
],
[
"Lee",
"Honglak",
""
],
[
"Orringer",
"Daniel A.",
""
]
] | TITLE: Artificial-intelligence-based molecular classification of diffuse
gliomas using rapid, label-free optical imaging
ABSTRACT: Molecular classification has transformed the management of brain tumors by
enabling more accurate prognostication and personalized treatment. However,
timely molecular diagnostic testing for patients with brain tumors is limited,
complicating surgical and adjuvant treatment and obstructing clinical trial
enrollment. In this study, we developed DeepGlioma, a rapid ($< 90$ seconds),
artificial-intelligence-based diagnostic screening system to streamline the
molecular diagnosis of diffuse gliomas. DeepGlioma is trained using a
multimodal dataset that includes stimulated Raman histology (SRH); a rapid,
label-free, non-consumptive, optical imaging method; and large-scale, public
genomic data. In a prospective, multicenter, international testing cohort of
patients with diffuse glioma ($n=153$) who underwent real-time SRH imaging, we
demonstrate that DeepGlioma can predict the molecular alterations used by the
World Health Organization to define the adult-type diffuse glioma taxonomy (IDH
mutation, 1p19q co-deletion and ATRX mutation), achieving a mean molecular
classification accuracy of $93.3\pm 1.6\%$. Our results represent how
artificial intelligence and optical histology can be used to provide a rapid
and scalable adjunct to wet lab methods for the molecular screening of patients
with diffuse glioma.
| new_dataset | 0.860017 |
2303.13627 | Mert Nak{\i}p | Erol Gelenbe and Mert Nak{\i}p | Associated Random Neural Networks for Collective Classification of Nodes
in Botnet Attacks | null | null | null | null | cs.NI cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Botnet attacks are a major threat to networked systems because of their
ability to turn the network nodes that they compromise into additional
attackers, leading to the spread of high volume attacks over long periods. The
detection of such Botnets is complicated by the fact that multiple network IP
addresses will be simultaneously compromised, so that Collective Classification
of compromised nodes, in addition to the already available traditional methods
that focus on individual nodes, can be useful. Thus this work introduces a
collective Botnet attack classification technique that operates on traffic from
an n-node IP network with a novel Associated Random Neural Network (ARNN) that
identifies the nodes which are compromised. The ARNN is a recurrent
architecture that incorporates two mutually associated, interconnected and
architecturally identical n-neuron random neural networks, that act
simultneously as mutual critics to reach the decision regarding which of n
nodes have been compromised. A novel gradient learning descent algorithm is
presented for the ARNN, and is shown to operate effectively both with
conventional off-line training from prior data, and with on-line incremental
training without prior off-line learning. Real data from a 107 node packet
network is used with over 700,000 packets to evaluate the ARNN, showing that it
provides accurate predictions. Comparisons with other well-known state of the
art methods using the same learning and testing datasets, show that the ARNN
offers significantly better performance.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 19:32:31 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Gelenbe",
"Erol",
""
],
[
"Nakıp",
"Mert",
""
]
] | TITLE: Associated Random Neural Networks for Collective Classification of Nodes
in Botnet Attacks
ABSTRACT: Botnet attacks are a major threat to networked systems because of their
ability to turn the network nodes that they compromise into additional
attackers, leading to the spread of high volume attacks over long periods. The
detection of such Botnets is complicated by the fact that multiple network IP
addresses will be simultaneously compromised, so that Collective Classification
of compromised nodes, in addition to the already available traditional methods
that focus on individual nodes, can be useful. Thus this work introduces a
collective Botnet attack classification technique that operates on traffic from
an n-node IP network with a novel Associated Random Neural Network (ARNN) that
identifies the nodes which are compromised. The ARNN is a recurrent
architecture that incorporates two mutually associated, interconnected and
architecturally identical n-neuron random neural networks, that act
simultneously as mutual critics to reach the decision regarding which of n
nodes have been compromised. A novel gradient learning descent algorithm is
presented for the ARNN, and is shown to operate effectively both with
conventional off-line training from prior data, and with on-line incremental
training without prior off-line learning. Real data from a 107 node packet
network is used with over 700,000 packets to evaluate the ARNN, showing that it
provides accurate predictions. Comparisons with other well-known state of the
art methods using the same learning and testing datasets, show that the ARNN
offers significantly better performance.
| no_new_dataset | 0.953275 |
2303.13636 | Yuntong Zhang | Yuntong Zhang, Jingye Xu, Mimi Xie, Wei Wang, Keying Ye, Jing Wang,
Dakai Zhu | PPG-based Heart Rate Estimation with Efficient Sensor Sampling and
Learning Models | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent studies showed that Photoplethysmography (PPG) sensors embedded in
wearable devices can estimate heart rate (HR) with high accuracy. However,
despite of prior research efforts, applying PPG sensor based HR estimation to
embedded devices still faces challenges due to the energy-intensive
high-frequency PPG sampling and the resource-intensive machine-learning models.
In this work, we aim to explore HR estimation techniques that are more suitable
for lower-power and resource-constrained embedded devices. More specifically,
we seek to design techniques that could provide high-accuracy HR estimation
with low-frequency PPG sampling, small model size, and fast inference time.
First, we show that by combining signal processing and ML, it is possible to
reduce the PPG sampling frequency from 125 Hz to only 25 Hz while providing
higher HR estimation accuracy. This combination also helps to reduce the ML
model feature size, leading to smaller models. Additionally, we present a
comprehensive analysis on different ML models and feature sizes to compare
their accuracy, model size, and inference time. The models explored include
Decision Tree (DT), Random Forest (RF), K-nearest neighbor (KNN), Support
vector machines (SVM), and Multi-layer perceptron (MLP). Experiments were
conducted using both a widely-utilized dataset and our self-collected dataset.
The experimental results show that our method by combining signal processing
and ML had only 5% error for HR estimation using low-frequency PPG data.
Moreover, our analysis showed that DT models with 10 to 20 input features
usually have good accuracy, while are several magnitude smaller in model sizes
and faster in inference time.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 19:47:36 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zhang",
"Yuntong",
""
],
[
"Xu",
"Jingye",
""
],
[
"Xie",
"Mimi",
""
],
[
"Wang",
"Wei",
""
],
[
"Ye",
"Keying",
""
],
[
"Wang",
"Jing",
""
],
[
"Zhu",
"Dakai",
""
]
] | TITLE: PPG-based Heart Rate Estimation with Efficient Sensor Sampling and
Learning Models
ABSTRACT: Recent studies showed that Photoplethysmography (PPG) sensors embedded in
wearable devices can estimate heart rate (HR) with high accuracy. However,
despite of prior research efforts, applying PPG sensor based HR estimation to
embedded devices still faces challenges due to the energy-intensive
high-frequency PPG sampling and the resource-intensive machine-learning models.
In this work, we aim to explore HR estimation techniques that are more suitable
for lower-power and resource-constrained embedded devices. More specifically,
we seek to design techniques that could provide high-accuracy HR estimation
with low-frequency PPG sampling, small model size, and fast inference time.
First, we show that by combining signal processing and ML, it is possible to
reduce the PPG sampling frequency from 125 Hz to only 25 Hz while providing
higher HR estimation accuracy. This combination also helps to reduce the ML
model feature size, leading to smaller models. Additionally, we present a
comprehensive analysis on different ML models and feature sizes to compare
their accuracy, model size, and inference time. The models explored include
Decision Tree (DT), Random Forest (RF), K-nearest neighbor (KNN), Support
vector machines (SVM), and Multi-layer perceptron (MLP). Experiments were
conducted using both a widely-utilized dataset and our self-collected dataset.
The experimental results show that our method by combining signal processing
and ML had only 5% error for HR estimation using low-frequency PPG data.
Moreover, our analysis showed that DT models with 10 to 20 input features
usually have good accuracy, while are several magnitude smaller in model sizes
and faster in inference time.
| no_new_dataset | 0.893263 |
2303.13637 | Yuntong Zhang | Yuntong Zhang, Jingye Xu, Mimi Xie, Dakai Zhu, Houbing Song, Wei Wang | Efficient and Direct Inference of Heart Rate Variability using Both
Signal Processing and Machine Learning | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Heart Rate Variability (HRV) measures the variation of the time between
consecutive heartbeats and is a major indicator of physical and mental health.
Recent research has demonstrated that photoplethysmography (PPG) sensors can be
used to infer HRV. However, many prior studies had high errors because they
only employed signal processing or machine learning (ML), or because they
indirectly inferred HRV, or because there lacks large training datasets. Many
prior studies may also require large ML models. The low accuracy and large
model sizes limit their applications to small embedded devices and potential
future use in healthcare. To address the above issues, we first collected a
large dataset of PPG signals and HRV ground truth. With this dataset, we
developed HRV models that combine signal processing and ML to directly infer
HRV. Evaluation results show that our method had errors between 3.5% to 25.7%
and outperformed signal-processing-only and ML-only methods. We also explored
different ML models, which showed that Decision Trees and Multi-level
Perceptrons have 13.0% and 9.1% errors on average with models at most hundreds
of KB and inference time less than 1ms. Hence, they are more suitable for small
embedded devices and potentially enable the future use of PPG-based HRV
monitoring in healthcare.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 19:47:53 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zhang",
"Yuntong",
""
],
[
"Xu",
"Jingye",
""
],
[
"Xie",
"Mimi",
""
],
[
"Zhu",
"Dakai",
""
],
[
"Song",
"Houbing",
""
],
[
"Wang",
"Wei",
""
]
] | TITLE: Efficient and Direct Inference of Heart Rate Variability using Both
Signal Processing and Machine Learning
ABSTRACT: Heart Rate Variability (HRV) measures the variation of the time between
consecutive heartbeats and is a major indicator of physical and mental health.
Recent research has demonstrated that photoplethysmography (PPG) sensors can be
used to infer HRV. However, many prior studies had high errors because they
only employed signal processing or machine learning (ML), or because they
indirectly inferred HRV, or because there lacks large training datasets. Many
prior studies may also require large ML models. The low accuracy and large
model sizes limit their applications to small embedded devices and potential
future use in healthcare. To address the above issues, we first collected a
large dataset of PPG signals and HRV ground truth. With this dataset, we
developed HRV models that combine signal processing and ML to directly infer
HRV. Evaluation results show that our method had errors between 3.5% to 25.7%
and outperformed signal-processing-only and ML-only methods. We also explored
different ML models, which showed that Decision Trees and Multi-level
Perceptrons have 13.0% and 9.1% errors on average with models at most hundreds
of KB and inference time less than 1ms. Hence, they are more suitable for small
embedded devices and potentially enable the future use of PPG-based HRV
monitoring in healthcare.
| new_dataset | 0.964119 |
2303.13648 | Wenxuan Wang | Haoran Wu, Wenxuan Wang, Yuxuan Wan, Wenxiang Jiao, Michael Lyu | ChatGPT or Grammarly? Evaluating ChatGPT on Grammatical Error Correction
Benchmark | Working in progress | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | ChatGPT is a cutting-edge artificial intelligence language model developed by
OpenAI, which has attracted a lot of attention due to its surprisingly strong
ability in answering follow-up questions. In this report, we aim to evaluate
ChatGPT on the Grammatical Error Correction(GEC) task, and compare it with
commercial GEC product (e.g., Grammarly) and state-of-the-art models (e.g.,
GECToR). By testing on the CoNLL2014 benchmark dataset, we find that ChatGPT
performs not as well as those baselines in terms of the automatic evaluation
metrics (e.g., $F_{0.5}$ score), particularly on long sentences. We inspect the
outputs and find that ChatGPT goes beyond one-by-one corrections. Specifically,
it prefers to change the surface expression of certain phrases or sentence
structure while maintaining grammatical correctness. Human evaluation
quantitatively confirms this and suggests that ChatGPT produces less
under-correction or mis-correction issues but more over-corrections. These
results demonstrate that ChatGPT is severely under-estimated by the automatic
evaluation metrics and could be a promising tool for GEC.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 00:35:50 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wu",
"Haoran",
""
],
[
"Wang",
"Wenxuan",
""
],
[
"Wan",
"Yuxuan",
""
],
[
"Jiao",
"Wenxiang",
""
],
[
"Lyu",
"Michael",
""
]
] | TITLE: ChatGPT or Grammarly? Evaluating ChatGPT on Grammatical Error Correction
Benchmark
ABSTRACT: ChatGPT is a cutting-edge artificial intelligence language model developed by
OpenAI, which has attracted a lot of attention due to its surprisingly strong
ability in answering follow-up questions. In this report, we aim to evaluate
ChatGPT on the Grammatical Error Correction(GEC) task, and compare it with
commercial GEC product (e.g., Grammarly) and state-of-the-art models (e.g.,
GECToR). By testing on the CoNLL2014 benchmark dataset, we find that ChatGPT
performs not as well as those baselines in terms of the automatic evaluation
metrics (e.g., $F_{0.5}$ score), particularly on long sentences. We inspect the
outputs and find that ChatGPT goes beyond one-by-one corrections. Specifically,
it prefers to change the surface expression of certain phrases or sentence
structure while maintaining grammatical correctness. Human evaluation
quantitatively confirms this and suggests that ChatGPT produces less
under-correction or mis-correction issues but more over-corrections. These
results demonstrate that ChatGPT is severely under-estimated by the automatic
evaluation metrics and could be a promising tool for GEC.
| no_new_dataset | 0.940681 |
2303.13653 | Monu Verma | Monu Verma, Murari Mandal, Satish Kumar Reddy, Yashwanth Reddy
Meedimale, Santosh Kumar Vipparthi | Efficient Neural Architecture Search for Emotion Recognition | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Automated human emotion recognition from facial expressions is a well-studied
problem and still remains a very challenging task. Some efficient or accurate
deep learning models have been presented in the literature. However, it is
quite difficult to design a model that is both efficient and accurate at the
same time. Moreover, identifying the minute feature variations in facial
regions for both macro and micro-expressions requires expertise in network
design. In this paper, we proposed to search for a highly efficient and robust
neural architecture for both macro and micro-level facial expression
recognition. To the best of our knowledge, this is the first attempt to design
a NAS-based solution for both macro and micro-expression recognition. We
produce lightweight models with a gradient-based architecture search algorithm.
To maintain consistency between macro and micro-expressions, we utilize dynamic
imaging and convert microexpression sequences into a single frame, preserving
the spatiotemporal features in the facial regions. The EmoNAS has evaluated
over 13 datasets (7 macro expression datasets: CK+, DISFA, MUG, ISED, OULU-VIS
CASIA, FER2013, RAF-DB, and 6 micro-expression datasets: CASME-I, CASME-II,
CAS(ME)2, SAMM, SMIC, MEGC2019 challenge). The proposed models outperform the
existing state-of-the-art methods and perform very well in terms of speed and
space complexity.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 20:21:26 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Verma",
"Monu",
""
],
[
"Mandal",
"Murari",
""
],
[
"Reddy",
"Satish Kumar",
""
],
[
"Meedimale",
"Yashwanth Reddy",
""
],
[
"Vipparthi",
"Santosh Kumar",
""
]
] | TITLE: Efficient Neural Architecture Search for Emotion Recognition
ABSTRACT: Automated human emotion recognition from facial expressions is a well-studied
problem and still remains a very challenging task. Some efficient or accurate
deep learning models have been presented in the literature. However, it is
quite difficult to design a model that is both efficient and accurate at the
same time. Moreover, identifying the minute feature variations in facial
regions for both macro and micro-expressions requires expertise in network
design. In this paper, we proposed to search for a highly efficient and robust
neural architecture for both macro and micro-level facial expression
recognition. To the best of our knowledge, this is the first attempt to design
a NAS-based solution for both macro and micro-expression recognition. We
produce lightweight models with a gradient-based architecture search algorithm.
To maintain consistency between macro and micro-expressions, we utilize dynamic
imaging and convert microexpression sequences into a single frame, preserving
the spatiotemporal features in the facial regions. The EmoNAS has evaluated
over 13 datasets (7 macro expression datasets: CK+, DISFA, MUG, ISED, OULU-VIS
CASIA, FER2013, RAF-DB, and 6 micro-expression datasets: CASME-I, CASME-II,
CAS(ME)2, SAMM, SMIC, MEGC2019 challenge). The proposed models outperform the
existing state-of-the-art methods and perform very well in terms of speed and
space complexity.
| no_new_dataset | 0.947624 |
2303.13662 | Yiyou Sun | Yiyou Sun, Yaojie Liu, Xiaoming Liu, Yixuan Li, Wen-Sheng Chu | Rethinking Domain Generalization for Face Anti-spoofing: Separability
and Alignment | Accepted in CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work studies the generalization issue of face anti-spoofing (FAS) models
on domain gaps, such as image resolution, blurriness and sensor variations.
Most prior works regard domain-specific signals as a negative impact, and apply
metric learning or adversarial losses to remove them from feature
representation. Though learning a domain-invariant feature space is viable for
the training data, we show that the feature shift still exists in an unseen
test domain, which backfires on the generalizability of the classifier. In this
work, instead of constructing a domain-invariant feature space, we encourage
domain separability while aligning the live-to-spoof transition (i.e., the
trajectory from live to spoof) to be the same for all domains. We formulate
this FAS strategy of separability and alignment (SA-FAS) as a problem of
invariant risk minimization (IRM), and learn domain-variant feature
representation but domain-invariant classifier. We demonstrate the
effectiveness of SA-FAS on challenging cross-domain FAS datasets and establish
state-of-the-art performance.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 20:34:27 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Sun",
"Yiyou",
""
],
[
"Liu",
"Yaojie",
""
],
[
"Liu",
"Xiaoming",
""
],
[
"Li",
"Yixuan",
""
],
[
"Chu",
"Wen-Sheng",
""
]
] | TITLE: Rethinking Domain Generalization for Face Anti-spoofing: Separability
and Alignment
ABSTRACT: This work studies the generalization issue of face anti-spoofing (FAS) models
on domain gaps, such as image resolution, blurriness and sensor variations.
Most prior works regard domain-specific signals as a negative impact, and apply
metric learning or adversarial losses to remove them from feature
representation. Though learning a domain-invariant feature space is viable for
the training data, we show that the feature shift still exists in an unseen
test domain, which backfires on the generalizability of the classifier. In this
work, instead of constructing a domain-invariant feature space, we encourage
domain separability while aligning the live-to-spoof transition (i.e., the
trajectory from live to spoof) to be the same for all domains. We formulate
this FAS strategy of separability and alignment (SA-FAS) as a problem of
invariant risk minimization (IRM), and learn domain-variant feature
representation but domain-invariant classifier. We demonstrate the
effectiveness of SA-FAS on challenging cross-domain FAS datasets and establish
state-of-the-art performance.
| no_new_dataset | 0.945551 |
2303.13664 | Anna Kukleva | Anna Kukleva, Moritz B\"ohle, Bernt Schiele, Hilde Kuehne, Christian
Rupprecht | Temperature Schedules for Self-Supervised Contrastive Methods on
Long-Tail Data | ICLR 2023 | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most approaches for self-supervised learning (SSL) are optimised on curated
balanced datasets, e.g. ImageNet, despite the fact that natural data usually
exhibits long-tail distributions. In this paper, we analyse the behaviour of
one of the most popular variants of SSL, i.e. contrastive methods, on long-tail
data. In particular, we investigate the role of the temperature parameter
$\tau$ in the contrastive loss, by analysing the loss through the lens of
average distance maximisation, and find that a large $\tau$ emphasises
group-wise discrimination, whereas a small $\tau$ leads to a higher degree of
instance discrimination. While $\tau$ has thus far been treated exclusively as
a constant hyperparameter, in this work, we propose to employ a dynamic $\tau$
and show that a simple cosine schedule can yield significant improvements in
the learnt representations. Such a schedule results in a constant `task
switching' between an emphasis on instance discrimination and group-wise
discrimination and thereby ensures that the model learns both group-wise
features, as well as instance-specific details. Since frequent classes benefit
from the former, while infrequent classes require the latter, we find this
method to consistently improve separation between the classes in long-tail data
without any additional computational cost.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 20:37:25 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Kukleva",
"Anna",
""
],
[
"Böhle",
"Moritz",
""
],
[
"Schiele",
"Bernt",
""
],
[
"Kuehne",
"Hilde",
""
],
[
"Rupprecht",
"Christian",
""
]
] | TITLE: Temperature Schedules for Self-Supervised Contrastive Methods on
Long-Tail Data
ABSTRACT: Most approaches for self-supervised learning (SSL) are optimised on curated
balanced datasets, e.g. ImageNet, despite the fact that natural data usually
exhibits long-tail distributions. In this paper, we analyse the behaviour of
one of the most popular variants of SSL, i.e. contrastive methods, on long-tail
data. In particular, we investigate the role of the temperature parameter
$\tau$ in the contrastive loss, by analysing the loss through the lens of
average distance maximisation, and find that a large $\tau$ emphasises
group-wise discrimination, whereas a small $\tau$ leads to a higher degree of
instance discrimination. While $\tau$ has thus far been treated exclusively as
a constant hyperparameter, in this work, we propose to employ a dynamic $\tau$
and show that a simple cosine schedule can yield significant improvements in
the learnt representations. Such a schedule results in a constant `task
switching' between an emphasis on instance discrimination and group-wise
discrimination and thereby ensures that the model learns both group-wise
features, as well as instance-specific details. Since frequent classes benefit
from the former, while infrequent classes require the latter, we find this
method to consistently improve separation between the classes in long-tail data
without any additional computational cost.
| no_new_dataset | 0.946101 |
2303.13714 | Troy Luhman | Troy Luhman, Eric Luhman | High Fidelity Image Synthesis With Deep VAEs In Latent Space | 19 pages, 16 figures | null | null | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | We present fast, realistic image generation on high-resolution, multimodal
datasets using hierarchical variational autoencoders (VAEs) trained on a
deterministic autoencoder's latent space. In this two-stage setup, the
autoencoder compresses the image into its semantic features, which are then
modeled with a deep VAE. With this method, the VAE avoids modeling the
fine-grained details that constitute the majority of the image's code length,
allowing it to focus on learning its structural components. We demonstrate the
effectiveness of our two-stage approach, achieving a FID of 9.34 on the
ImageNet-256 dataset which is comparable to BigGAN. We make our implementation
available online.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 23:45:19 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Luhman",
"Troy",
""
],
[
"Luhman",
"Eric",
""
]
] | TITLE: High Fidelity Image Synthesis With Deep VAEs In Latent Space
ABSTRACT: We present fast, realistic image generation on high-resolution, multimodal
datasets using hierarchical variational autoencoders (VAEs) trained on a
deterministic autoencoder's latent space. In this two-stage setup, the
autoencoder compresses the image into its semantic features, which are then
modeled with a deep VAE. With this method, the VAE avoids modeling the
fine-grained details that constitute the majority of the image's code length,
allowing it to focus on learning its structural components. We demonstrate the
effectiveness of our two-stage approach, achieving a FID of 9.34 on the
ImageNet-256 dataset which is comparable to BigGAN. We make our implementation
available online.
| no_new_dataset | 0.946597 |
2303.13724 | Weide Liu | Weide Liu, Zhonghua Wu, Yang Zhao, Yuming Fang, Chuan-Sheng Foo, Jun
Cheng and Guosheng Lin | Harmonizing Base and Novel Classes: A Class-Contrastive Approach for
Generalized Few-Shot Segmentation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current methods for few-shot segmentation (FSSeg) have mainly focused on
improving the performance of novel classes while neglecting the performance of
base classes. To overcome this limitation, the task of generalized few-shot
semantic segmentation (GFSSeg) has been introduced, aiming to predict
segmentation masks for both base and novel classes. However, the current
prototype-based methods do not explicitly consider the relationship between
base and novel classes when updating prototypes, leading to a limited
performance in identifying true categories. To address this challenge, we
propose a class contrastive loss and a class relationship loss to regulate
prototype updates and encourage a large distance between prototypes from
different classes, thus distinguishing the classes from each other while
maintaining the performance of the base classes. Our proposed approach achieves
new state-of-the-art performance for the generalized few-shot segmentation task
on PASCAL VOC and MS COCO datasets.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 00:30:25 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Liu",
"Weide",
""
],
[
"Wu",
"Zhonghua",
""
],
[
"Zhao",
"Yang",
""
],
[
"Fang",
"Yuming",
""
],
[
"Foo",
"Chuan-Sheng",
""
],
[
"Cheng",
"Jun",
""
],
[
"Lin",
"Guosheng",
""
]
] | TITLE: Harmonizing Base and Novel Classes: A Class-Contrastive Approach for
Generalized Few-Shot Segmentation
ABSTRACT: Current methods for few-shot segmentation (FSSeg) have mainly focused on
improving the performance of novel classes while neglecting the performance of
base classes. To overcome this limitation, the task of generalized few-shot
semantic segmentation (GFSSeg) has been introduced, aiming to predict
segmentation masks for both base and novel classes. However, the current
prototype-based methods do not explicitly consider the relationship between
base and novel classes when updating prototypes, leading to a limited
performance in identifying true categories. To address this challenge, we
propose a class contrastive loss and a class relationship loss to regulate
prototype updates and encourage a large distance between prototypes from
different classes, thus distinguishing the classes from each other while
maintaining the performance of the base classes. Our proposed approach achieves
new state-of-the-art performance for the generalized few-shot segmentation task
on PASCAL VOC and MS COCO datasets.
| no_new_dataset | 0.949995 |
2303.13735 | Junyu Chen | Junyu Chen, Norihiro Yoshida, Hiroaki Takada | An investigation of licensing of datasets for machine learning based on
the GQM model | null | null | null | null | cs.SE cs.CY cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dataset licensing is currently an issue in the development of machine
learning systems. And in the development of machine learning systems, the most
widely used are publicly available datasets. However, since the images in the
publicly available dataset are mainly obtained from the Internet, some images
are not commercially available. Furthermore, developers of machine learning
systems do not often care about the license of the dataset when training
machine learning models with it. In summary, the licensing of datasets for
machine learning systems is in a state of incompleteness in all aspects at this
stage.
Our investigation of two collection datasets revealed that most of the
current datasets lacked licenses, and the lack of licenses made it impossible
to determine the commercial availability of the datasets. Therefore, we decided
to take a more scientific and systematic approach to investigate the licensing
of datasets and the licensing of machine learning systems that use the dataset
to make it easier and more compliant for future developers of machine learning
systems.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 01:39:56 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Chen",
"Junyu",
""
],
[
"Yoshida",
"Norihiro",
""
],
[
"Takada",
"Hiroaki",
""
]
] | TITLE: An investigation of licensing of datasets for machine learning based on
the GQM model
ABSTRACT: Dataset licensing is currently an issue in the development of machine
learning systems. And in the development of machine learning systems, the most
widely used are publicly available datasets. However, since the images in the
publicly available dataset are mainly obtained from the Internet, some images
are not commercially available. Furthermore, developers of machine learning
systems do not often care about the license of the dataset when training
machine learning models with it. In summary, the licensing of datasets for
machine learning systems is in a state of incompleteness in all aspects at this
stage.
Our investigation of two collection datasets revealed that most of the
current datasets lacked licenses, and the lack of licenses made it impossible
to determine the commercial availability of the datasets. Therefore, we decided
to take a more scientific and systematic approach to investigate the licensing
of datasets and the licensing of machine learning systems that use the dataset
to make it easier and more compliant for future developers of machine learning
systems.
| no_new_dataset | 0.951369 |
2303.13744 | Haomiao Ni | Haomiao Ni, Changhao Shi, Kai Li, Sharon X. Huang, Martin Renqiang Min | Conditional Image-to-Video Generation with Latent Flow Diffusion Models | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conditional image-to-video (cI2V) generation aims to synthesize a new
plausible video starting from an image (e.g., a person's face) and a condition
(e.g., an action class label like smile). The key challenge of the cI2V task
lies in the simultaneous generation of realistic spatial appearance and
temporal dynamics corresponding to the given image and condition. In this
paper, we propose an approach for cI2V using novel latent flow diffusion models
(LFDM) that synthesize an optical flow sequence in the latent space based on
the given condition to warp the given image. Compared to previous
direct-synthesis-based works, our proposed LFDM can better synthesize spatial
details and temporal motion by fully utilizing the spatial content of the given
image and warping it in the latent space according to the generated
temporally-coherent flow. The training of LFDM consists of two separate stages:
(1) an unsupervised learning stage to train a latent flow auto-encoder for
spatial content generation, including a flow predictor to estimate latent flow
between pairs of video frames, and (2) a conditional learning stage to train a
3D-UNet-based diffusion model (DM) for temporal latent flow generation. Unlike
previous DMs operating in pixel space or latent feature space that couples
spatial and temporal information, the DM in our LFDM only needs to learn a
low-dimensional latent flow space for motion generation, thus being more
computationally efficient. We conduct comprehensive experiments on multiple
datasets, where LFDM consistently outperforms prior arts. Furthermore, we show
that LFDM can be easily adapted to new domains by simply finetuning the image
decoder. Our code is available at https://github.com/nihaomiao/CVPR23_LFDM.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 01:54:26 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ni",
"Haomiao",
""
],
[
"Shi",
"Changhao",
""
],
[
"Li",
"Kai",
""
],
[
"Huang",
"Sharon X.",
""
],
[
"Min",
"Martin Renqiang",
""
]
] | TITLE: Conditional Image-to-Video Generation with Latent Flow Diffusion Models
ABSTRACT: Conditional image-to-video (cI2V) generation aims to synthesize a new
plausible video starting from an image (e.g., a person's face) and a condition
(e.g., an action class label like smile). The key challenge of the cI2V task
lies in the simultaneous generation of realistic spatial appearance and
temporal dynamics corresponding to the given image and condition. In this
paper, we propose an approach for cI2V using novel latent flow diffusion models
(LFDM) that synthesize an optical flow sequence in the latent space based on
the given condition to warp the given image. Compared to previous
direct-synthesis-based works, our proposed LFDM can better synthesize spatial
details and temporal motion by fully utilizing the spatial content of the given
image and warping it in the latent space according to the generated
temporally-coherent flow. The training of LFDM consists of two separate stages:
(1) an unsupervised learning stage to train a latent flow auto-encoder for
spatial content generation, including a flow predictor to estimate latent flow
between pairs of video frames, and (2) a conditional learning stage to train a
3D-UNet-based diffusion model (DM) for temporal latent flow generation. Unlike
previous DMs operating in pixel space or latent feature space that couples
spatial and temporal information, the DM in our LFDM only needs to learn a
low-dimensional latent flow space for motion generation, thus being more
computationally efficient. We conduct comprehensive experiments on multiple
datasets, where LFDM consistently outperforms prior arts. Furthermore, we show
that LFDM can be easily adapted to new domains by simply finetuning the image
decoder. Our code is available at https://github.com/nihaomiao/CVPR23_LFDM.
| no_new_dataset | 0.953144 |
2303.13753 | Meiqi Hu | Meiqi Hu, Chen Wu, Bo Du | EMS-Net: Efficient Multi-Temporal Self-Attention For Hyperspectral
Change Detection | 4 pages, 5 figures, submitted to IGARSS2023 | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by-sa/4.0/ | Hyperspectral change detection plays an essential role of monitoring the
dynamic urban development and detecting precise fine object evolution and
alteration. In this paper, we have proposed an original Efficient
Multi-temporal Self-attention Network (EMS-Net) for hyperspectral change
detection. The designed EMS module cuts redundancy of those similar and
containing-no-changes feature maps, computing efficient multi-temporal change
information for precise binary change map. Besides, to explore the clustering
characteristics of the change detection, a novel supervised contrastive loss is
provided to enhance the compactness of the unchanged. Experiments implemented
on two hyperspectral change detection datasets manifests the out-standing
performance and validity of proposed method.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 02:11:22 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Hu",
"Meiqi",
""
],
[
"Wu",
"Chen",
""
],
[
"Du",
"Bo",
""
]
] | TITLE: EMS-Net: Efficient Multi-Temporal Self-Attention For Hyperspectral
Change Detection
ABSTRACT: Hyperspectral change detection plays an essential role of monitoring the
dynamic urban development and detecting precise fine object evolution and
alteration. In this paper, we have proposed an original Efficient
Multi-temporal Self-attention Network (EMS-Net) for hyperspectral change
detection. The designed EMS module cuts redundancy of those similar and
containing-no-changes feature maps, computing efficient multi-temporal change
information for precise binary change map. Besides, to explore the clustering
characteristics of the change detection, a novel supervised contrastive loss is
provided to enhance the compactness of the unchanged. Experiments implemented
on two hyperspectral change detection datasets manifests the out-standing
performance and validity of proposed method.
| no_new_dataset | 0.949106 |
2303.13756 | Zhenyu Xie | Zhenyu Xie and Zaiyu Huang and Xin Dong and Fuwei Zhao and Haoye Dong
and Xijin Zhang and Feida Zhu and Xiaodan Liang | GP-VTON: Towards General Purpose Virtual Try-on via Collaborative
Local-Flow Global-Parsing Learning | 8 pages, 8 figures, The IEEE/CVF Computer Vision and Pattern
Recognition Conference (CVPR) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image-based Virtual Try-ON aims to transfer an in-shop garment onto a
specific person. Existing methods employ a global warping module to model the
anisotropic deformation for different garment parts, which fails to preserve
the semantic information of different parts when receiving challenging inputs
(e.g, intricate human poses, difficult garments). Moreover, most of them
directly warp the input garment to align with the boundary of the preserved
region, which usually requires texture squeezing to meet the boundary shape
constraint and thus leads to texture distortion. The above inferior performance
hinders existing methods from real-world applications. To address these
problems and take a step towards real-world virtual try-on, we propose a
General-Purpose Virtual Try-ON framework, named GP-VTON, by developing an
innovative Local-Flow Global-Parsing (LFGP) warping module and a Dynamic
Gradient Truncation (DGT) training strategy. Specifically, compared with the
previous global warping mechanism, LFGP employs local flows to warp garments
parts individually, and assembles the local warped results via the global
garment parsing, resulting in reasonable warped parts and a semantic-correct
intact garment even with challenging inputs.On the other hand, our DGT training
strategy dynamically truncates the gradient in the overlap area and the warped
garment is no more required to meet the boundary constraint, which effectively
avoids the texture squeezing problem. Furthermore, our GP-VTON can be easily
extended to multi-category scenario and jointly trained by using data from
different garment categories. Extensive experiments on two high-resolution
benchmarks demonstrate our superiority over the existing state-of-the-art
methods.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 02:12:29 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Xie",
"Zhenyu",
""
],
[
"Huang",
"Zaiyu",
""
],
[
"Dong",
"Xin",
""
],
[
"Zhao",
"Fuwei",
""
],
[
"Dong",
"Haoye",
""
],
[
"Zhang",
"Xijin",
""
],
[
"Zhu",
"Feida",
""
],
[
"Liang",
"Xiaodan",
""
]
] | TITLE: GP-VTON: Towards General Purpose Virtual Try-on via Collaborative
Local-Flow Global-Parsing Learning
ABSTRACT: Image-based Virtual Try-ON aims to transfer an in-shop garment onto a
specific person. Existing methods employ a global warping module to model the
anisotropic deformation for different garment parts, which fails to preserve
the semantic information of different parts when receiving challenging inputs
(e.g, intricate human poses, difficult garments). Moreover, most of them
directly warp the input garment to align with the boundary of the preserved
region, which usually requires texture squeezing to meet the boundary shape
constraint and thus leads to texture distortion. The above inferior performance
hinders existing methods from real-world applications. To address these
problems and take a step towards real-world virtual try-on, we propose a
General-Purpose Virtual Try-ON framework, named GP-VTON, by developing an
innovative Local-Flow Global-Parsing (LFGP) warping module and a Dynamic
Gradient Truncation (DGT) training strategy. Specifically, compared with the
previous global warping mechanism, LFGP employs local flows to warp garments
parts individually, and assembles the local warped results via the global
garment parsing, resulting in reasonable warped parts and a semantic-correct
intact garment even with challenging inputs.On the other hand, our DGT training
strategy dynamically truncates the gradient in the overlap area and the warped
garment is no more required to meet the boundary constraint, which effectively
avoids the texture squeezing problem. Furthermore, our GP-VTON can be easily
extended to multi-category scenario and jointly trained by using data from
different garment categories. Extensive experiments on two high-resolution
benchmarks demonstrate our superiority over the existing state-of-the-art
methods.
| no_new_dataset | 0.951097 |
2303.13777 | Jianchuan Chen | Jianchuan Chen, Wentao Yi, Liqian Ma, Xu Jia, Huchuan Lu | GM-NeRF: Learning Generalizable Model-based Neural Radiance Fields from
Multi-view Images | Accepted at CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we focus on synthesizing high-fidelity novel view images for
arbitrary human performers, given a set of sparse multi-view images. It is a
challenging task due to the large variation among articulated body poses and
heavy self-occlusions. To alleviate this, we introduce an effective
generalizable framework Generalizable Model-based Neural Radiance Fields
(GM-NeRF) to synthesize free-viewpoint images. Specifically, we propose a
geometry-guided attention mechanism to register the appearance code from
multi-view 2D images to a geometry proxy which can alleviate the misalignment
between inaccurate geometry prior and pixel space. On top of that, we further
conduct neural rendering and partial gradient backpropagation for efficient
perceptual supervision and improvement of the perceptual quality of synthesis.
To evaluate our method, we conduct experiments on synthesized datasets
THuman2.0 and Multi-garment, and real-world datasets Genebody and ZJUMocap. The
results demonstrate that our approach outperforms state-of-the-art methods in
terms of novel view synthesis and geometric reconstruction.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 03:32:02 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Chen",
"Jianchuan",
""
],
[
"Yi",
"Wentao",
""
],
[
"Ma",
"Liqian",
""
],
[
"Jia",
"Xu",
""
],
[
"Lu",
"Huchuan",
""
]
] | TITLE: GM-NeRF: Learning Generalizable Model-based Neural Radiance Fields from
Multi-view Images
ABSTRACT: In this work, we focus on synthesizing high-fidelity novel view images for
arbitrary human performers, given a set of sparse multi-view images. It is a
challenging task due to the large variation among articulated body poses and
heavy self-occlusions. To alleviate this, we introduce an effective
generalizable framework Generalizable Model-based Neural Radiance Fields
(GM-NeRF) to synthesize free-viewpoint images. Specifically, we propose a
geometry-guided attention mechanism to register the appearance code from
multi-view 2D images to a geometry proxy which can alleviate the misalignment
between inaccurate geometry prior and pixel space. On top of that, we further
conduct neural rendering and partial gradient backpropagation for efficient
perceptual supervision and improvement of the perceptual quality of synthesis.
To evaluate our method, we conduct experiments on synthesized datasets
THuman2.0 and Multi-garment, and real-world datasets Genebody and ZJUMocap. The
results demonstrate that our approach outperforms state-of-the-art methods in
terms of novel view synthesis and geometric reconstruction.
| no_new_dataset | 0.946349 |
2303.13788 | Minjie Hua | Minjie Hua, Yibing Nan, Shiguo Lian | Application-Driven AI Paradigm for Person Counting in Various Scenarios | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Person counting is considered as a fundamental task in video surveillance.
However, the scenario diversity in practical applications makes it difficult to
exploit a single person counting model for general use. Consequently, engineers
must preview the video stream and manually specify an appropriate person
counting model based on the scenario of camera shot, which is time-consuming,
especially for large-scale deployments. In this paper, we propose a person
counting paradigm that utilizes a scenario classifier to automatically select a
suitable person counting model for each captured frame. First, the input image
is passed through the scenario classifier to obtain a scenario label, which is
then used to allocate the frame to one of five fine-tuned models for person
counting. Additionally, we present five augmentation datasets collected from
different scenarios, including side-view, long-shot, top-view, customized and
crowd, which are also integrated to form a scenario classification dataset
containing 26323 samples. In our comparative experiments, the proposed paradigm
achieves better balance than any single model on the integrated dataset, thus
its generalization in various scenarios has been proved.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 03:57:21 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Hua",
"Minjie",
""
],
[
"Nan",
"Yibing",
""
],
[
"Lian",
"Shiguo",
""
]
] | TITLE: Application-Driven AI Paradigm for Person Counting in Various Scenarios
ABSTRACT: Person counting is considered as a fundamental task in video surveillance.
However, the scenario diversity in practical applications makes it difficult to
exploit a single person counting model for general use. Consequently, engineers
must preview the video stream and manually specify an appropriate person
counting model based on the scenario of camera shot, which is time-consuming,
especially for large-scale deployments. In this paper, we propose a person
counting paradigm that utilizes a scenario classifier to automatically select a
suitable person counting model for each captured frame. First, the input image
is passed through the scenario classifier to obtain a scenario label, which is
then used to allocate the frame to one of five fine-tuned models for person
counting. Additionally, we present five augmentation datasets collected from
different scenarios, including side-view, long-shot, top-view, customized and
crowd, which are also integrated to form a scenario classification dataset
containing 26323 samples. In our comparative experiments, the proposed paradigm
achieves better balance than any single model on the integrated dataset, thus
its generalization in various scenarios has been proved.
| new_dataset | 0.958265 |
2303.13791 | Andreas Meuleman | Andreas Meuleman and Yu-Lun Liu and Chen Gao and Jia-Bin Huang and
Changil Kim and Min H. Kim and Johannes Kopf | Progressively Optimized Local Radiance Fields for Robust View Synthesis | Project page: https://localrf.github.io/ | CVPR 2023 | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present an algorithm for reconstructing the radiance field of a
large-scale scene from a single casually captured video. The task poses two
core challenges. First, most existing radiance field reconstruction approaches
rely on accurate pre-estimated camera poses from Structure-from-Motion
algorithms, which frequently fail on in-the-wild videos. Second, using a
single, global radiance field with finite representational capacity does not
scale to longer trajectories in an unbounded scene. For handling unknown poses,
we jointly estimate the camera poses with radiance field in a progressive
manner. We show that progressive optimization significantly improves the
robustness of the reconstruction. For handling large unbounded scenes, we
dynamically allocate new local radiance fields trained with frames within a
temporal window. This further improves robustness (e.g., performs well even
under moderate pose drifts) and allows us to scale to large scenes. Our
extensive evaluation on the Tanks and Temples dataset and our collected outdoor
dataset, Static Hikes, show that our approach compares favorably with the
state-of-the-art.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 04:03:55 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Meuleman",
"Andreas",
""
],
[
"Liu",
"Yu-Lun",
""
],
[
"Gao",
"Chen",
""
],
[
"Huang",
"Jia-Bin",
""
],
[
"Kim",
"Changil",
""
],
[
"Kim",
"Min H.",
""
],
[
"Kopf",
"Johannes",
""
]
] | TITLE: Progressively Optimized Local Radiance Fields for Robust View Synthesis
ABSTRACT: We present an algorithm for reconstructing the radiance field of a
large-scale scene from a single casually captured video. The task poses two
core challenges. First, most existing radiance field reconstruction approaches
rely on accurate pre-estimated camera poses from Structure-from-Motion
algorithms, which frequently fail on in-the-wild videos. Second, using a
single, global radiance field with finite representational capacity does not
scale to longer trajectories in an unbounded scene. For handling unknown poses,
we jointly estimate the camera poses with radiance field in a progressive
manner. We show that progressive optimization significantly improves the
robustness of the reconstruction. For handling large unbounded scenes, we
dynamically allocate new local radiance fields trained with frames within a
temporal window. This further improves robustness (e.g., performs well even
under moderate pose drifts) and allows us to scale to large scenes. Our
extensive evaluation on the Tanks and Temples dataset and our collected outdoor
dataset, Static Hikes, show that our approach compares favorably with the
state-of-the-art.
| no_new_dataset | 0.756358 |
2303.13794 | Hongjian Song | Hongjian Song, Yuki Kashiwaba, Shuai Wu, Canming Wang | Efficient and Accurate Co-Visible Region Localization with Matching
Key-Points Crop (MKPC): A Two-Stage Pipeline for Enhancing Image Matching
Performance | 9 pages with 6 figures. Many experiments have not yet been conducted,
the theoretical sections are rather concise, and the references are not
adequately comprehensive. This version of the paper is being released to make
this work public, and code will also be published soon. We will continue to
conduct additional experiments and periodically update the paper | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Image matching is a classic and fundamental task in computer vision. In this
paper, under the hypothesis that the areas outside the co-visible regions carry
little information, we propose a matching key-points crop (MKPC) algorithm. The
MKPC locates, proposes and crops the critical regions, which are the co-visible
areas with great efficiency and accuracy. Furthermore, building upon MKPC, we
propose a general two-stage pipeline for image matching, which is compatible to
any image matching models or combinations. We experimented with plugging
SuperPoint + SuperGlue into the two-stage pipeline, whose results show that our
method enhances the performance for outdoor pose estimations. What's more, in a
fair comparative condition, our method outperforms the SOTA on Image Matching
Challenge 2022 Benchmark, which represents the hardest outdoor benchmark of
image matching currently.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 04:18:13 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Song",
"Hongjian",
""
],
[
"Kashiwaba",
"Yuki",
""
],
[
"Wu",
"Shuai",
""
],
[
"Wang",
"Canming",
""
]
] | TITLE: Efficient and Accurate Co-Visible Region Localization with Matching
Key-Points Crop (MKPC): A Two-Stage Pipeline for Enhancing Image Matching
Performance
ABSTRACT: Image matching is a classic and fundamental task in computer vision. In this
paper, under the hypothesis that the areas outside the co-visible regions carry
little information, we propose a matching key-points crop (MKPC) algorithm. The
MKPC locates, proposes and crops the critical regions, which are the co-visible
areas with great efficiency and accuracy. Furthermore, building upon MKPC, we
propose a general two-stage pipeline for image matching, which is compatible to
any image matching models or combinations. We experimented with plugging
SuperPoint + SuperGlue into the two-stage pipeline, whose results show that our
method enhances the performance for outdoor pose estimations. What's more, in a
fair comparative condition, our method outperforms the SOTA on Image Matching
Challenge 2022 Benchmark, which represents the hardest outdoor benchmark of
image matching currently.
| no_new_dataset | 0.94474 |
2303.13797 | A.B. Siddique | A.B. Siddique, M.H. Maqbool, Kshitija Taywade, Hassan Foroosh | Personalizing Task-oriented Dialog Systems via Zero-shot Generalizable
Reward Function | 11 pages, 4 tables, 31st ACM International Conference on Information
and Knowledge Management (CIKM'22) | null | 10.1145/3511808.3557417 | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Task-oriented dialog systems enable users to accomplish tasks using natural
language. State-of-the-art systems respond to users in the same way regardless
of their personalities, although personalizing dialogues can lead to higher
levels of adoption and better user experiences. Building personalized dialog
systems is an important, yet challenging endeavor and only a handful of works
took on the challenge. Most existing works rely on supervised learning
approaches and require laborious and expensive labeled training data for each
user profile. Additionally, collecting and labeling data for each user profile
is virtually impossible. In this work, we propose a novel framework, P-ToD, to
personalize task-oriented dialog systems capable of adapting to a wide range of
user profiles in an unsupervised fashion using a zero-shot generalizable reward
function. P-ToD uses a pre-trained GPT-2 as a backbone model and works in three
phases. Phase one performs task-specific training. Phase two kicks off
unsupervised personalization by leveraging the proximal policy optimization
algorithm that performs policy gradients guided by the zero-shot generalizable
reward function. Our novel reward function can quantify the quality of the
generated responses even for unseen profiles. The optional final phase
fine-tunes the personalized model using a few labeled training examples. We
conduct extensive experimental analysis using the personalized bAbI dialogue
benchmark for five tasks and up to 180 diverse user profiles. The experimental
results demonstrate that P-ToD, even when it had access to zero labeled
examples, outperforms state-of-the-art supervised personalization models and
achieves competitive performance on BLEU and ROUGE metrics when compared to a
strong fully-supervised GPT-2 baseline
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 04:33:40 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Siddique",
"A. B.",
""
],
[
"Maqbool",
"M. H.",
""
],
[
"Taywade",
"Kshitija",
""
],
[
"Foroosh",
"Hassan",
""
]
] | TITLE: Personalizing Task-oriented Dialog Systems via Zero-shot Generalizable
Reward Function
ABSTRACT: Task-oriented dialog systems enable users to accomplish tasks using natural
language. State-of-the-art systems respond to users in the same way regardless
of their personalities, although personalizing dialogues can lead to higher
levels of adoption and better user experiences. Building personalized dialog
systems is an important, yet challenging endeavor and only a handful of works
took on the challenge. Most existing works rely on supervised learning
approaches and require laborious and expensive labeled training data for each
user profile. Additionally, collecting and labeling data for each user profile
is virtually impossible. In this work, we propose a novel framework, P-ToD, to
personalize task-oriented dialog systems capable of adapting to a wide range of
user profiles in an unsupervised fashion using a zero-shot generalizable reward
function. P-ToD uses a pre-trained GPT-2 as a backbone model and works in three
phases. Phase one performs task-specific training. Phase two kicks off
unsupervised personalization by leveraging the proximal policy optimization
algorithm that performs policy gradients guided by the zero-shot generalizable
reward function. Our novel reward function can quantify the quality of the
generated responses even for unseen profiles. The optional final phase
fine-tunes the personalized model using a few labeled training examples. We
conduct extensive experimental analysis using the personalized bAbI dialogue
benchmark for five tasks and up to 180 diverse user profiles. The experimental
results demonstrate that P-ToD, even when it had access to zero labeled
examples, outperforms state-of-the-art supervised personalization models and
achieves competitive performance on BLEU and ROUGE metrics when compared to a
strong fully-supervised GPT-2 baseline
| no_new_dataset | 0.948155 |
2303.13798 | Mohammadreza Sharif | Mohammadreza Sharif, Kiran Mohan, Sarath Suvarna | 2D Floor Plan Segmentation Based on Down-sampling | null | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by/4.0/ | In recent years, floor plan segmentation has gained significant attention due
to its wide range of applications in floor plan reconstruction and robotics. In
this paper, we propose a novel 2D floor plan segmentation technique based on a
down-sampling approach. Our method employs continuous down-sampling on a floor
plan to maintain its structural information while reducing its complexity. We
demonstrate the effectiveness of our approach by presenting results obtained
from both cluttered floor plans generated by a vacuum cleaning robot in unknown
environments and a benchmark of floor plans. Our technique considerably reduces
the computational and implementation complexity of floor plan segmentation,
making it more suitable for real-world applications. Additionally, we discuss
the appropriate metric for evaluating segmentation results. Overall, our
approach yields promising results for 2D floor plan segmentation in cluttered
environments.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 04:39:50 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Sharif",
"Mohammadreza",
""
],
[
"Mohan",
"Kiran",
""
],
[
"Suvarna",
"Sarath",
""
]
] | TITLE: 2D Floor Plan Segmentation Based on Down-sampling
ABSTRACT: In recent years, floor plan segmentation has gained significant attention due
to its wide range of applications in floor plan reconstruction and robotics. In
this paper, we propose a novel 2D floor plan segmentation technique based on a
down-sampling approach. Our method employs continuous down-sampling on a floor
plan to maintain its structural information while reducing its complexity. We
demonstrate the effectiveness of our approach by presenting results obtained
from both cluttered floor plans generated by a vacuum cleaning robot in unknown
environments and a benchmark of floor plans. Our technique considerably reduces
the computational and implementation complexity of floor plan segmentation,
making it more suitable for real-world applications. Additionally, we discuss
the appropriate metric for evaluating segmentation results. Overall, our
approach yields promising results for 2D floor plan segmentation in cluttered
environments.
| no_new_dataset | 0.957397 |
2303.13801 | A.B. Siddique | Adib Mosharrof, Moghis Fereidouni, A.B. Siddique | Toward Open-domain Slot Filling via Self-supervised Co-training | 10 pages, 6 tables, 2 figures, ACM Web Conference 2023 (WWW'23) | null | 10.1145/3543507.3583541 | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Slot filling is one of the critical tasks in modern conversational systems.
The majority of existing literature employs supervised learning methods, which
require labeled training data for each new domain. Zero-shot learning and weak
supervision approaches, among others, have shown promise as alternatives to
manual labeling. Nonetheless, these learning paradigms are significantly
inferior to supervised learning approaches in terms of performance. To minimize
this performance gap and demonstrate the possibility of open-domain slot
filling, we propose a Self-supervised Co-training framework, called SCot, that
requires zero in-domain manually labeled training examples and works in three
phases. Phase one acquires two sets of complementary pseudo labels
automatically. Phase two leverages the power of the pre-trained language model
BERT, by adapting it for the slot filling task using these sets of pseudo
labels. In phase three, we introduce a self-supervised cotraining mechanism,
where both models automatically select highconfidence soft labels to further
improve the performance of the other in an iterative fashion. Our thorough
evaluations show that SCot outperforms state-of-the-art models by 45.57% and
37.56% on SGD and MultiWoZ datasets, respectively. Moreover, our proposed
framework SCot achieves comparable performance when compared to
state-of-the-art fully supervised models.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 04:51:22 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Mosharrof",
"Adib",
""
],
[
"Fereidouni",
"Moghis",
""
],
[
"Siddique",
"A. B.",
""
]
] | TITLE: Toward Open-domain Slot Filling via Self-supervised Co-training
ABSTRACT: Slot filling is one of the critical tasks in modern conversational systems.
The majority of existing literature employs supervised learning methods, which
require labeled training data for each new domain. Zero-shot learning and weak
supervision approaches, among others, have shown promise as alternatives to
manual labeling. Nonetheless, these learning paradigms are significantly
inferior to supervised learning approaches in terms of performance. To minimize
this performance gap and demonstrate the possibility of open-domain slot
filling, we propose a Self-supervised Co-training framework, called SCot, that
requires zero in-domain manually labeled training examples and works in three
phases. Phase one acquires two sets of complementary pseudo labels
automatically. Phase two leverages the power of the pre-trained language model
BERT, by adapting it for the slot filling task using these sets of pseudo
labels. In phase three, we introduce a self-supervised cotraining mechanism,
where both models automatically select highconfidence soft labels to further
improve the performance of the other in an iterative fashion. Our thorough
evaluations show that SCot outperforms state-of-the-art models by 45.57% and
37.56% on SGD and MultiWoZ datasets, respectively. Moreover, our proposed
framework SCot achieves comparable performance when compared to
state-of-the-art fully supervised models.
| no_new_dataset | 0.949576 |
2303.13810 | Xinwen Liu | Xinwen Liu, Jing Wang, S. Kevin Zhou, Craig Engstrom, Shekhar S.
Chandra | Evidence-aware multi-modal data fusion and its application to total knee
replacement prediction | null | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural networks have been widely studied for predicting a medical
condition, such as total knee replacement (TKR). It has shown that data of
different modalities, such as imaging data, clinical variables and demographic
information, provide complementary information and thus can improve the
prediction accuracy together. However, the data sources of various modalities
may not always be of high quality, and each modality may have only partial
information of medical condition. Thus, predictions from different modalities
can be opposite, and the final prediction may fail in the presence of such a
conflict. Therefore, it is important to consider the reliability of each source
data and the prediction output when making a final decision. In this paper, we
propose an evidence-aware multi-modal data fusion framework based on the
Dempster-Shafer theory (DST). The backbone models contain an image branch, a
non-image branch and a fusion branch. For each branch, there is an evidence
network that takes the extracted features as input and outputs an evidence
score, which is designed to represent the reliability of the output from the
current branch. The output probabilities along with the evidence scores from
multiple branches are combined with the Dempster's combination rule to make a
final prediction. Experimental results on the public OA initiative (OAI)
dataset for the TKR prediction task show the superiority of the proposed fusion
strategy on various backbone models.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 05:06:02 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Liu",
"Xinwen",
""
],
[
"Wang",
"Jing",
""
],
[
"Zhou",
"S. Kevin",
""
],
[
"Engstrom",
"Craig",
""
],
[
"Chandra",
"Shekhar S.",
""
]
] | TITLE: Evidence-aware multi-modal data fusion and its application to total knee
replacement prediction
ABSTRACT: Deep neural networks have been widely studied for predicting a medical
condition, such as total knee replacement (TKR). It has shown that data of
different modalities, such as imaging data, clinical variables and demographic
information, provide complementary information and thus can improve the
prediction accuracy together. However, the data sources of various modalities
may not always be of high quality, and each modality may have only partial
information of medical condition. Thus, predictions from different modalities
can be opposite, and the final prediction may fail in the presence of such a
conflict. Therefore, it is important to consider the reliability of each source
data and the prediction output when making a final decision. In this paper, we
propose an evidence-aware multi-modal data fusion framework based on the
Dempster-Shafer theory (DST). The backbone models contain an image branch, a
non-image branch and a fusion branch. For each branch, there is an evidence
network that takes the extracted features as input and outputs an evidence
score, which is designed to represent the reliability of the output from the
current branch. The output probabilities along with the evidence scores from
multiple branches are combined with the Dempster's combination rule to make a
final prediction. Experimental results on the public OA initiative (OAI)
dataset for the TKR prediction task show the superiority of the proposed fusion
strategy on various backbone models.
| no_new_dataset | 0.954308 |
2303.13825 | Zhiyang Guo | Zhiyang Guo, Wengang Zhou, Min Wang, Li Li, Houqiang Li | HandNeRF: Neural Radiance Fields for Animatable Interacting Hands | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a novel framework to reconstruct accurate appearance and geometry
with neural radiance fields (NeRF) for interacting hands, enabling the
rendering of photo-realistic images and videos for gesture animation from
arbitrary views. Given multi-view images of a single hand or interacting hands,
an off-the-shelf skeleton estimator is first employed to parameterize the hand
poses. Then we design a pose-driven deformation field to establish
correspondence from those different poses to a shared canonical space, where a
pose-disentangled NeRF for one hand is optimized. Such unified modeling
efficiently complements the geometry and texture cues in rarely-observed areas
for both hands. Meanwhile, we further leverage the pose priors to generate
pseudo depth maps as guidance for occlusion-aware density learning. Moreover, a
neural feature distillation method is proposed to achieve cross-domain
alignment for color optimization. We conduct extensive experiments to verify
the merits of our proposed HandNeRF and report a series of state-of-the-art
results both qualitatively and quantitatively on the large-scale InterHand2.6M
dataset.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 06:19:19 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Guo",
"Zhiyang",
""
],
[
"Zhou",
"Wengang",
""
],
[
"Wang",
"Min",
""
],
[
"Li",
"Li",
""
],
[
"Li",
"Houqiang",
""
]
] | TITLE: HandNeRF: Neural Radiance Fields for Animatable Interacting Hands
ABSTRACT: We propose a novel framework to reconstruct accurate appearance and geometry
with neural radiance fields (NeRF) for interacting hands, enabling the
rendering of photo-realistic images and videos for gesture animation from
arbitrary views. Given multi-view images of a single hand or interacting hands,
an off-the-shelf skeleton estimator is first employed to parameterize the hand
poses. Then we design a pose-driven deformation field to establish
correspondence from those different poses to a shared canonical space, where a
pose-disentangled NeRF for one hand is optimized. Such unified modeling
efficiently complements the geometry and texture cues in rarely-observed areas
for both hands. Meanwhile, we further leverage the pose priors to generate
pseudo depth maps as guidance for occlusion-aware density learning. Moreover, a
neural feature distillation method is proposed to achieve cross-domain
alignment for color optimization. We conduct extensive experiments to verify
the merits of our proposed HandNeRF and report a series of state-of-the-art
results both qualitatively and quantitatively on the large-scale InterHand2.6M
dataset.
| no_new_dataset | 0.949295 |
2303.13839 | Jiefeng Ma | Jiefeng Ma, Jun Du, Pengfei Hu, Zhenrong Zhang, Jianshu Zhang, Huihui
Zhu, Cong Liu | HRDoc: Dataset and Baseline Method Toward Hierarchical Reconstruction of
Document Structures | 8 pages, 6 figures. Accepted by AAAI-2023 | null | null | null | cs.CL cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The problem of document structure reconstruction refers to converting digital
or scanned documents into corresponding semantic structures. Most existing
works mainly focus on splitting the boundary of each element in a single
document page, neglecting the reconstruction of semantic structure in
multi-page documents. This paper introduces hierarchical reconstruction of
document structures as a novel task suitable for NLP and CV fields. To better
evaluate the system performance on the new task, we built a large-scale dataset
named HRDoc, which consists of 2,500 multi-page documents with nearly 2 million
semantic units. Every document in HRDoc has line-level annotations including
categories and relations obtained from rule-based extractors and human
annotators. Moreover, we proposed an encoder-decoder-based hierarchical
document structure parsing system (DSPS) to tackle this problem. By adopting a
multi-modal bidirectional encoder and a structure-aware GRU decoder with
soft-mask operation, the DSPS model surpass the baseline method by a large
margin. All scripts and datasets will be made publicly available at
https://github.com/jfma-USTC/HRDoc.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 07:23:56 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ma",
"Jiefeng",
""
],
[
"Du",
"Jun",
""
],
[
"Hu",
"Pengfei",
""
],
[
"Zhang",
"Zhenrong",
""
],
[
"Zhang",
"Jianshu",
""
],
[
"Zhu",
"Huihui",
""
],
[
"Liu",
"Cong",
""
]
] | TITLE: HRDoc: Dataset and Baseline Method Toward Hierarchical Reconstruction of
Document Structures
ABSTRACT: The problem of document structure reconstruction refers to converting digital
or scanned documents into corresponding semantic structures. Most existing
works mainly focus on splitting the boundary of each element in a single
document page, neglecting the reconstruction of semantic structure in
multi-page documents. This paper introduces hierarchical reconstruction of
document structures as a novel task suitable for NLP and CV fields. To better
evaluate the system performance on the new task, we built a large-scale dataset
named HRDoc, which consists of 2,500 multi-page documents with nearly 2 million
semantic units. Every document in HRDoc has line-level annotations including
categories and relations obtained from rule-based extractors and human
annotators. Moreover, we proposed an encoder-decoder-based hierarchical
document structure parsing system (DSPS) to tackle this problem. By adopting a
multi-modal bidirectional encoder and a structure-aware GRU decoder with
soft-mask operation, the DSPS model surpass the baseline method by a large
margin. All scripts and datasets will be made publicly available at
https://github.com/jfma-USTC/HRDoc.
| new_dataset | 0.962143 |
2303.13844 | Yue Pang | Lei Zou, Yue Pang, M. Tamer \"Ozsu, Jiaqi Chen | Efficient Execution of SPARQL Queries with OPTIONAL and UNION
Expressions | null | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The proliferation of RDF datasets has resulted in studies focusing on
optimizing SPARQL query processing. Most existing work focuses on basic graph
patterns (BGPs) and ignores other vital operators in SPARQL, such as UNION and
OPTIONAL. SPARQL queries with these operators, which we abbreviate as
SPARQL-UO, pose serious query plan generation challenges. In this paper, we
propose techniques for executing SPARQL-UO queries using BGP execution as a
building block, based on a novel BGP-based Evaluation (BE)-Tree representation
of query plans. On top of this, we propose a series of cost-driven BE-tree
transformations to generate more efficient plans by reducing the search space
and intermediate result sizes, and a candidate pruning technique that further
enhances efficiency at query time. Experiments confirm that our method
outperforms the state-of-the-art by orders of magnitude.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 07:37:11 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zou",
"Lei",
""
],
[
"Pang",
"Yue",
""
],
[
"Özsu",
"M. Tamer",
""
],
[
"Chen",
"Jiaqi",
""
]
] | TITLE: Efficient Execution of SPARQL Queries with OPTIONAL and UNION
Expressions
ABSTRACT: The proliferation of RDF datasets has resulted in studies focusing on
optimizing SPARQL query processing. Most existing work focuses on basic graph
patterns (BGPs) and ignores other vital operators in SPARQL, such as UNION and
OPTIONAL. SPARQL queries with these operators, which we abbreviate as
SPARQL-UO, pose serious query plan generation challenges. In this paper, we
propose techniques for executing SPARQL-UO queries using BGP execution as a
building block, based on a novel BGP-based Evaluation (BE)-Tree representation
of query plans. On top of this, we propose a series of cost-driven BE-tree
transformations to generate more efficient plans by reducing the search space
and intermediate result sizes, and a candidate pruning technique that further
enhances efficiency at query time. Experiments confirm that our method
outperforms the state-of-the-art by orders of magnitude.
| no_new_dataset | 0.947039 |
2303.13852 | Renjiao Yi | Renjiao Yi, Chenyang Zhu, Kai Xu | Weakly-supervised Single-view Image Relighting | 21 pages, with supplementary material | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a learning-based approach to relight a single image of Lambertian
and low-frequency specular objects. Our method enables inserting objects from
photographs into new scenes and relighting them under the new environment
lighting, which is essential for AR applications. To relight the object, we
solve both inverse rendering and re-rendering. To resolve the ill-posed inverse
rendering, we propose a weakly-supervised method by a low-rank constraint. To
facilitate the weakly-supervised training, we contribute Relit, a large-scale
(750K images) dataset of videos with aligned objects under changing
illuminations. For re-rendering, we propose a differentiable specular rendering
layer to render low-frequency non-Lambertian materials under various
illuminations of spherical harmonics. The whole pipeline is end-to-end and
efficient, allowing for a mobile app implementation of AR object insertion.
Extensive evaluations demonstrate that our method achieves state-of-the-art
performance. Project page: https://renjiaoyi.github.io/relighting/.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 08:20:16 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Yi",
"Renjiao",
""
],
[
"Zhu",
"Chenyang",
""
],
[
"Xu",
"Kai",
""
]
] | TITLE: Weakly-supervised Single-view Image Relighting
ABSTRACT: We present a learning-based approach to relight a single image of Lambertian
and low-frequency specular objects. Our method enables inserting objects from
photographs into new scenes and relighting them under the new environment
lighting, which is essential for AR applications. To relight the object, we
solve both inverse rendering and re-rendering. To resolve the ill-posed inverse
rendering, we propose a weakly-supervised method by a low-rank constraint. To
facilitate the weakly-supervised training, we contribute Relit, a large-scale
(750K images) dataset of videos with aligned objects under changing
illuminations. For re-rendering, we propose a differentiable specular rendering
layer to render low-frequency non-Lambertian materials under various
illuminations of spherical harmonics. The whole pipeline is end-to-end and
efficient, allowing for a mobile app implementation of AR object insertion.
Extensive evaluations demonstrate that our method achieves state-of-the-art
performance. Project page: https://renjiaoyi.github.io/relighting/.
| new_dataset | 0.954223 |
2303.13853 | Mikhail Kennerley | Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, Robby T. Tan | 2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised
Domain Adaptive Object Detection | Accepted into CVPR'23 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Object detection at night is a challenging problem due to the absence of
night image annotations. Despite several domain adaptation methods, achieving
high-precision results remains an issue. False-positive error propagation is
still observed in methods using the well-established student-teacher framework,
particularly for small-scale and low-light objects. This paper proposes a
two-phase consistency unsupervised domain adaptation network, 2PCNet, to
address these issues. The network employs high-confidence bounding-box
predictions from the teacher in the first phase and appends them to the
student's region proposals for the teacher to re-evaluate in the second phase,
resulting in a combination of high and low confidence pseudo-labels. The night
images and pseudo-labels are scaled-down before being used as input to the
student, providing stronger small-scale pseudo-labels. To address errors that
arise from low-light regions and other night-related attributes in images, we
propose a night-specific augmentation pipeline called NightAug. This pipeline
involves applying random augmentations, such as glare, blur, and noise, to
daytime images. Experiments on publicly available datasets demonstrate that our
method achieves superior results to state-of-the-art methods by 20\%, and to
supervised models trained directly on the target data.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 08:22:41 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Kennerley",
"Mikhail",
""
],
[
"Wang",
"Jian-Gang",
""
],
[
"Veeravalli",
"Bharadwaj",
""
],
[
"Tan",
"Robby T.",
""
]
] | TITLE: 2PCNet: Two-Phase Consistency Training for Day-to-Night Unsupervised
Domain Adaptive Object Detection
ABSTRACT: Object detection at night is a challenging problem due to the absence of
night image annotations. Despite several domain adaptation methods, achieving
high-precision results remains an issue. False-positive error propagation is
still observed in methods using the well-established student-teacher framework,
particularly for small-scale and low-light objects. This paper proposes a
two-phase consistency unsupervised domain adaptation network, 2PCNet, to
address these issues. The network employs high-confidence bounding-box
predictions from the teacher in the first phase and appends them to the
student's region proposals for the teacher to re-evaluate in the second phase,
resulting in a combination of high and low confidence pseudo-labels. The night
images and pseudo-labels are scaled-down before being used as input to the
student, providing stronger small-scale pseudo-labels. To address errors that
arise from low-light regions and other night-related attributes in images, we
propose a night-specific augmentation pipeline called NightAug. This pipeline
involves applying random augmentations, such as glare, blur, and noise, to
daytime images. Experiments on publicly available datasets demonstrate that our
method achieves superior results to state-of-the-art methods by 20\%, and to
supervised models trained directly on the target data.
| no_new_dataset | 0.948298 |
2303.13862 | Hao Chen | Hao Chen, Linyan Li, Fan Lyu, Fuyuan Hu, Zhenping Xia and Fenglei Xu | Two-level Graph Network for Few-Shot Class-Incremental Learning | arXiv admin note: text overlap with arXiv:2203.06953 by other authors | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Few-shot class-incremental learning (FSCIL) aims to design machine learning
algorithms that can continually learn new concepts from a few data points,
without forgetting knowledge of old classes. The difficulty lies in that
limited data from new classes not only lead to significant overfitting issues
but also exacerbates the notorious catastrophic forgetting problems. However,
existing FSCIL methods ignore the semantic relationships between sample-level
and class-level. % Using the advantage that graph neural network (GNN) can mine
rich information among few samples, In this paper, we designed a two-level
graph network for FSCIL named Sample-level and Class-level Graph Neural Network
(SCGN). Specifically, a pseudo incremental learning paradigm is designed in
SCGN, which synthesizes virtual few-shot tasks as new tasks to optimize SCGN
model parameters in advance. Sample-level graph network uses the relationship
of a few samples to aggregate similar samples and obtains refined class-level
features. Class-level graph network aims to mitigate the semantic conflict
between prototype features of new classes and old classes. SCGN builds
two-level graph networks to guarantee the latent semantic of each few-shot
class can be effectively represented in FSCIL. Experiments on three popular
benchmark datasets show that our method significantly outperforms the baselines
and sets new state-of-the-art results with remarkable advantages.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 08:58:08 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Chen",
"Hao",
""
],
[
"Li",
"Linyan",
""
],
[
"Lyu",
"Fan",
""
],
[
"Hu",
"Fuyuan",
""
],
[
"Xia",
"Zhenping",
""
],
[
"Xu",
"Fenglei",
""
]
] | TITLE: Two-level Graph Network for Few-Shot Class-Incremental Learning
ABSTRACT: Few-shot class-incremental learning (FSCIL) aims to design machine learning
algorithms that can continually learn new concepts from a few data points,
without forgetting knowledge of old classes. The difficulty lies in that
limited data from new classes not only lead to significant overfitting issues
but also exacerbates the notorious catastrophic forgetting problems. However,
existing FSCIL methods ignore the semantic relationships between sample-level
and class-level. % Using the advantage that graph neural network (GNN) can mine
rich information among few samples, In this paper, we designed a two-level
graph network for FSCIL named Sample-level and Class-level Graph Neural Network
(SCGN). Specifically, a pseudo incremental learning paradigm is designed in
SCGN, which synthesizes virtual few-shot tasks as new tasks to optimize SCGN
model parameters in advance. Sample-level graph network uses the relationship
of a few samples to aggregate similar samples and obtains refined class-level
features. Class-level graph network aims to mitigate the semantic conflict
between prototype features of new classes and old classes. SCGN builds
two-level graph networks to guarantee the latent semantic of each few-shot
class can be effectively represented in FSCIL. Experiments on three popular
benchmark datasets show that our method significantly outperforms the baselines
and sets new state-of-the-art results with remarkable advantages.
| no_new_dataset | 0.953101 |
2303.13874 | WonJun Moon | WonJun Moon, Sangeek Hyun, SangUk Park, Dongchan Park, Jae-Pil Heo | Query-Dependent Video Representation for Moment Retrieval and Highlight
Detection | Accepted to CVPR 2023. Code is available at
https://github.com/wjun0830/QD-DETR | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | Recently, video moment retrieval and highlight detection (MR/HD) are being
spotlighted as the demand for video understanding is drastically increased. The
key objective of MR/HD is to localize the moment and estimate clip-wise
accordance level, i.e., saliency score, to the given text query. Although the
recent transformer-based models brought some advances, we found that these
methods do not fully exploit the information of a given query. For example, the
relevance between text query and video contents is sometimes neglected when
predicting the moment and its saliency. To tackle this issue, we introduce
Query-Dependent DETR (QD-DETR), a detection transformer tailored for MR/HD. As
we observe the insignificant role of a given query in transformer
architectures, our encoding module starts with cross-attention layers to
explicitly inject the context of text query into video representation. Then, to
enhance the model's capability of exploiting the query information, we
manipulate the video-query pairs to produce irrelevant pairs. Such negative
(irrelevant) video-query pairs are trained to yield low saliency scores, which
in turn, encourages the model to estimate precise accordance between
query-video pairs. Lastly, we present an input-adaptive saliency predictor
which adaptively defines the criterion of saliency scores for the given
video-query pairs. Our extensive studies verify the importance of building the
query-dependent representation for MR/HD. Specifically, QD-DETR outperforms
state-of-the-art methods on QVHighlights, TVSum, and Charades-STA datasets.
Codes are available at github.com/wjun0830/QD-DETR.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 09:32:50 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Moon",
"WonJun",
""
],
[
"Hyun",
"Sangeek",
""
],
[
"Park",
"SangUk",
""
],
[
"Park",
"Dongchan",
""
],
[
"Heo",
"Jae-Pil",
""
]
] | TITLE: Query-Dependent Video Representation for Moment Retrieval and Highlight
Detection
ABSTRACT: Recently, video moment retrieval and highlight detection (MR/HD) are being
spotlighted as the demand for video understanding is drastically increased. The
key objective of MR/HD is to localize the moment and estimate clip-wise
accordance level, i.e., saliency score, to the given text query. Although the
recent transformer-based models brought some advances, we found that these
methods do not fully exploit the information of a given query. For example, the
relevance between text query and video contents is sometimes neglected when
predicting the moment and its saliency. To tackle this issue, we introduce
Query-Dependent DETR (QD-DETR), a detection transformer tailored for MR/HD. As
we observe the insignificant role of a given query in transformer
architectures, our encoding module starts with cross-attention layers to
explicitly inject the context of text query into video representation. Then, to
enhance the model's capability of exploiting the query information, we
manipulate the video-query pairs to produce irrelevant pairs. Such negative
(irrelevant) video-query pairs are trained to yield low saliency scores, which
in turn, encourages the model to estimate precise accordance between
query-video pairs. Lastly, we present an input-adaptive saliency predictor
which adaptively defines the criterion of saliency scores for the given
video-query pairs. Our extensive studies verify the importance of building the
query-dependent representation for MR/HD. Specifically, QD-DETR outperforms
state-of-the-art methods on QVHighlights, TVSum, and Charades-STA datasets.
Codes are available at github.com/wjun0830/QD-DETR.
| no_new_dataset | 0.943348 |
2303.13881 | Carlos Hernandez-Olivan | Carlos Hernandez-Olivan, Sonia Rubio Llamas, Jose R. Beltran | Symbolic Music Structure Analysis with Graph Representations and
Changepoint Detection Methods | null | null | null | null | cs.SD cs.AI eess.AS | http://creativecommons.org/licenses/by-sa/4.0/ | Music Structure Analysis is an open research task in Music Information
Retrieval (MIR). In the past, there have been several works that attempt to
segment music into the audio and symbolic domains, however, the identification
and segmentation of the music structure at different levels is still an open
research problem in this area. In this work we propose three methods, two of
which are novel graph-based algorithms that aim to segment symbolic music by
its form or structure: Norm, G-PELT and G-Window. We performed an ablation
study with two public datasets that have different forms or structures in order
to compare such methods varying their parameter values and comparing the
performance against different music styles. We have found that encoding
symbolic music with graph representations and computing the novelty of
Adjacency Matrices obtained from graphs represent the structure of symbolic
music pieces well without the need to extract features from it. We are able to
detect the boundaries with an online unsupervised changepoint detection method
with a F_1 of 0.5640 for a 1 bar tolerance in one of the public datasets that
we used for testing our methods. We also provide the performance results of the
algorithms at different levels of structure, high, medium and low, to show how
the parameters of the proposed methods have to be adjusted depending on the
level. We added the best performing method with its parameters for each
structure level to musicaiz, an open source python package, to facilitate the
reproducibility and usability of this work. We hope that this methods could be
used to improve other MIR tasks such as music generation with structure, music
classification or key changes detection.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 09:45:11 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Hernandez-Olivan",
"Carlos",
""
],
[
"Llamas",
"Sonia Rubio",
""
],
[
"Beltran",
"Jose R.",
""
]
] | TITLE: Symbolic Music Structure Analysis with Graph Representations and
Changepoint Detection Methods
ABSTRACT: Music Structure Analysis is an open research task in Music Information
Retrieval (MIR). In the past, there have been several works that attempt to
segment music into the audio and symbolic domains, however, the identification
and segmentation of the music structure at different levels is still an open
research problem in this area. In this work we propose three methods, two of
which are novel graph-based algorithms that aim to segment symbolic music by
its form or structure: Norm, G-PELT and G-Window. We performed an ablation
study with two public datasets that have different forms or structures in order
to compare such methods varying their parameter values and comparing the
performance against different music styles. We have found that encoding
symbolic music with graph representations and computing the novelty of
Adjacency Matrices obtained from graphs represent the structure of symbolic
music pieces well without the need to extract features from it. We are able to
detect the boundaries with an online unsupervised changepoint detection method
with a F_1 of 0.5640 for a 1 bar tolerance in one of the public datasets that
we used for testing our methods. We also provide the performance results of the
algorithms at different levels of structure, high, medium and low, to show how
the parameters of the proposed methods have to be adjusted depending on the
level. We added the best performing method with its parameters for each
structure level to musicaiz, an open source python package, to facilitate the
reproducibility and usability of this work. We hope that this methods could be
used to improve other MIR tasks such as music generation with structure, music
classification or key changes detection.
| no_new_dataset | 0.945147 |
2303.13885 | Junsong Chen | Haojie Zhao and Junsong Chen and Lijun Wang and Huchuan Lu | ARKitTrack: A New Diverse Dataset for Tracking Using Mobile RGB-D Data | Accepted by CVPR2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Compared with traditional RGB-only visual tracking, few datasets have been
constructed for RGB-D tracking. In this paper, we propose ARKitTrack, a new
RGB-D tracking dataset for both static and dynamic scenes captured by
consumer-grade LiDAR scanners equipped on Apple's iPhone and iPad. ARKitTrack
contains 300 RGB-D sequences, 455 targets, and 229.7K video frames in total.
Along with the bounding box annotations and frame-level attributes, we also
annotate this dataset with 123.9K pixel-level target masks. Besides, the camera
intrinsic and camera pose of each frame are provided for future developments.
To demonstrate the potential usefulness of this dataset, we further present a
unified baseline for both box-level and pixel-level tracking, which integrates
RGB features with bird's-eye-view representations to better explore
cross-modality 3D geometry. In-depth empirical analysis has verified that the
ARKitTrack dataset can significantly facilitate RGB-D tracking and that the
proposed baseline method compares favorably against the state of the arts. The
code and dataset is available at https://arkittrack.github.io.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 09:51:13 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zhao",
"Haojie",
""
],
[
"Chen",
"Junsong",
""
],
[
"Wang",
"Lijun",
""
],
[
"Lu",
"Huchuan",
""
]
] | TITLE: ARKitTrack: A New Diverse Dataset for Tracking Using Mobile RGB-D Data
ABSTRACT: Compared with traditional RGB-only visual tracking, few datasets have been
constructed for RGB-D tracking. In this paper, we propose ARKitTrack, a new
RGB-D tracking dataset for both static and dynamic scenes captured by
consumer-grade LiDAR scanners equipped on Apple's iPhone and iPad. ARKitTrack
contains 300 RGB-D sequences, 455 targets, and 229.7K video frames in total.
Along with the bounding box annotations and frame-level attributes, we also
annotate this dataset with 123.9K pixel-level target masks. Besides, the camera
intrinsic and camera pose of each frame are provided for future developments.
To demonstrate the potential usefulness of this dataset, we further present a
unified baseline for both box-level and pixel-level tracking, which integrates
RGB features with bird's-eye-view representations to better explore
cross-modality 3D geometry. In-depth empirical analysis has verified that the
ARKitTrack dataset can significantly facilitate RGB-D tracking and that the
proposed baseline method compares favorably against the state of the arts. The
code and dataset is available at https://arkittrack.github.io.
| new_dataset | 0.961353 |
2303.13896 | Grigorios Chrysos | Grigorios G Chrysos, Bohan Wang, Jiankang Deng, Volkan Cevher | Regularization of polynomial networks for image recognition | Accepted at CVPR'23 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Deep Neural Networks (DNNs) have obtained impressive performance across
tasks, however they still remain as black boxes, e.g., hard to theoretically
analyze. At the same time, Polynomial Networks (PNs) have emerged as an
alternative method with a promising performance and improved interpretability
but have yet to reach the performance of the powerful DNN baselines. In this
work, we aim to close this performance gap. We introduce a class of PNs, which
are able to reach the performance of ResNet across a range of six benchmarks.
We demonstrate that strong regularization is critical and conduct an extensive
study of the exact regularization schemes required to match performance. To
further motivate the regularization schemes, we introduce D-PolyNets that
achieve a higher-degree of expansion than previously proposed polynomial
networks. D-PolyNets are more parameter-efficient while achieving a similar
performance as other polynomial networks. We expect that our new models can
lead to an understanding of the role of elementwise activation functions (which
are no longer required for training PNs). The source code is available at
https://github.com/grigorisg9gr/regularized_polynomials.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 10:05:22 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Chrysos",
"Grigorios G",
""
],
[
"Wang",
"Bohan",
""
],
[
"Deng",
"Jiankang",
""
],
[
"Cevher",
"Volkan",
""
]
] | TITLE: Regularization of polynomial networks for image recognition
ABSTRACT: Deep Neural Networks (DNNs) have obtained impressive performance across
tasks, however they still remain as black boxes, e.g., hard to theoretically
analyze. At the same time, Polynomial Networks (PNs) have emerged as an
alternative method with a promising performance and improved interpretability
but have yet to reach the performance of the powerful DNN baselines. In this
work, we aim to close this performance gap. We introduce a class of PNs, which
are able to reach the performance of ResNet across a range of six benchmarks.
We demonstrate that strong regularization is critical and conduct an extensive
study of the exact regularization schemes required to match performance. To
further motivate the regularization schemes, we introduce D-PolyNets that
achieve a higher-degree of expansion than previously proposed polynomial
networks. D-PolyNets are more parameter-efficient while achieving a similar
performance as other polynomial networks. We expect that our new models can
lead to an understanding of the role of elementwise activation functions (which
are no longer required for training PNs). The source code is available at
https://github.com/grigorisg9gr/regularized_polynomials.
| no_new_dataset | 0.941868 |
2303.13898 | Zhiheng Ma | Zhiheng Ma, Xiaopeng Hong, Beinan Liu, Yabin Wang, Pinyue Guo, Huiyun
Li | Remind of the Past: Incremental Learning with Analogical Prompts | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although data-free incremental learning methods are memory-friendly,
accurately estimating and counteracting representation shifts is challenging in
the absence of historical data. This paper addresses this thorny problem by
proposing a novel incremental learning method inspired by human analogy
capabilities. Specifically, we design an analogy-making mechanism to remap the
new data into the old class by prompt tuning. It mimics the feature
distribution of the target old class on the old model using only samples of new
classes. The learnt prompts are further used to estimate and counteract the
representation shift caused by fine-tuning for the historical prototypes. The
proposed method sets up new state-of-the-art performance on four incremental
learning benchmarks under both the class and domain incremental learning
settings. It consistently outperforms data-replay methods by only saving
feature prototypes for each class. It has almost hit the empirical upper bound
by joint training on the Core50 benchmark. The code will be released at
\url{https://github.com/ZhihengCV/A-Prompts}.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 10:18:28 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ma",
"Zhiheng",
""
],
[
"Hong",
"Xiaopeng",
""
],
[
"Liu",
"Beinan",
""
],
[
"Wang",
"Yabin",
""
],
[
"Guo",
"Pinyue",
""
],
[
"Li",
"Huiyun",
""
]
] | TITLE: Remind of the Past: Incremental Learning with Analogical Prompts
ABSTRACT: Although data-free incremental learning methods are memory-friendly,
accurately estimating and counteracting representation shifts is challenging in
the absence of historical data. This paper addresses this thorny problem by
proposing a novel incremental learning method inspired by human analogy
capabilities. Specifically, we design an analogy-making mechanism to remap the
new data into the old class by prompt tuning. It mimics the feature
distribution of the target old class on the old model using only samples of new
classes. The learnt prompts are further used to estimate and counteract the
representation shift caused by fine-tuning for the historical prototypes. The
proposed method sets up new state-of-the-art performance on four incremental
learning benchmarks under both the class and domain incremental learning
settings. It consistently outperforms data-replay methods by only saving
feature prototypes for each class. It has almost hit the empirical upper bound
by joint training on the Core50 benchmark. The code will be released at
\url{https://github.com/ZhihengCV/A-Prompts}.
| no_new_dataset | 0.946349 |
2303.13916 | Junji Otsuka | Junji Otsuka, Masakazu Yoshimura, Takeshi Ohashi | Self-Supervised Reversed Image Signal Processing via Reference-Guided
Dynamic Parameter Selection | 19 pages, 12 figures | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Unprocessed sensor outputs (RAW images) potentially improve both low-level
and high-level computer vision algorithms, but the lack of large-scale RAW
image datasets is a barrier to research. Thus, reversed Image Signal Processing
(ISP) which converts existing RGB images into RAW images has been studied.
However, most existing methods require camera-specific metadata or paired RGB
and RAW images to model the conversion, and they are not always available. In
addition, there are issues in handling diverse ISPs and recovering global
illumination. To tackle these limitations, we propose a self-supervised
reversed ISP method that does not require metadata and paired images. The
proposed method converts a RGB image into a RAW-like image taken in the same
environment with the same sensor as a reference RAW image by dynamically
selecting parameters of the reversed ISP pipeline based on the reference RAW
image. The parameter selection is trained via pseudo paired data created from
unpaired RGB and RAW images. We show that the proposed method is able to learn
various reversed ISPs with comparable accuracy to other state-of-the-art
supervised methods and convert unknown RGB images from COCO and Flickr1M to
target RAW-like images more accurately in terms of pixel distribution. We also
demonstrate that our generated RAW images improve performance on real RAW image
object detection task.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 11:12:05 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Otsuka",
"Junji",
""
],
[
"Yoshimura",
"Masakazu",
""
],
[
"Ohashi",
"Takeshi",
""
]
] | TITLE: Self-Supervised Reversed Image Signal Processing via Reference-Guided
Dynamic Parameter Selection
ABSTRACT: Unprocessed sensor outputs (RAW images) potentially improve both low-level
and high-level computer vision algorithms, but the lack of large-scale RAW
image datasets is a barrier to research. Thus, reversed Image Signal Processing
(ISP) which converts existing RGB images into RAW images has been studied.
However, most existing methods require camera-specific metadata or paired RGB
and RAW images to model the conversion, and they are not always available. In
addition, there are issues in handling diverse ISPs and recovering global
illumination. To tackle these limitations, we propose a self-supervised
reversed ISP method that does not require metadata and paired images. The
proposed method converts a RGB image into a RAW-like image taken in the same
environment with the same sensor as a reference RAW image by dynamically
selecting parameters of the reversed ISP pipeline based on the reference RAW
image. The parameter selection is trained via pseudo paired data created from
unpaired RGB and RAW images. We show that the proposed method is able to learn
various reversed ISPs with comparable accuracy to other state-of-the-art
supervised methods and convert unknown RGB images from COCO and Flickr1M to
target RAW-like images more accurately in terms of pixel distribution. We also
demonstrate that our generated RAW images improve performance on real RAW image
object detection task.
| no_new_dataset | 0.955486 |
2303.13931 | Yousri Kessentini | Marwa Dhiaf, Ahmed Cheikh Rouhou, Yousri Kessentini, Sinda Ben Salem | MSdocTr-Lite: A Lite Transformer for Full Page Multi-script Handwriting
Recognition | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The Transformer has quickly become the dominant architecture for various
pattern recognition tasks due to its capacity for long-range representation.
However, transformers are data-hungry models and need large datasets for
training. In Handwritten Text Recognition (HTR), collecting a massive amount of
labeled data is a complicated and expensive task. In this paper, we propose a
lite transformer architecture for full-page multi-script handwriting
recognition. The proposed model comes with three advantages: First, to solve
the common problem of data scarcity, we propose a lite transformer model that
can be trained on a reasonable amount of data, which is the case of most HTR
public datasets, without the need for external data. Second, it can learn the
reading order at page-level thanks to a curriculum learning strategy, allowing
it to avoid line segmentation errors, exploit a larger context and reduce the
need for costly segmentation annotations. Third, it can be easily adapted to
other scripts by applying a simple transfer-learning process using only
page-level labeled images. Extensive experiments on different datasets with
different scripts (French, English, Spanish, and Arabic) show the effectiveness
of the proposed model.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 11:40:50 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Dhiaf",
"Marwa",
""
],
[
"Rouhou",
"Ahmed Cheikh",
""
],
[
"Kessentini",
"Yousri",
""
],
[
"Salem",
"Sinda Ben",
""
]
] | TITLE: MSdocTr-Lite: A Lite Transformer for Full Page Multi-script Handwriting
Recognition
ABSTRACT: The Transformer has quickly become the dominant architecture for various
pattern recognition tasks due to its capacity for long-range representation.
However, transformers are data-hungry models and need large datasets for
training. In Handwritten Text Recognition (HTR), collecting a massive amount of
labeled data is a complicated and expensive task. In this paper, we propose a
lite transformer architecture for full-page multi-script handwriting
recognition. The proposed model comes with three advantages: First, to solve
the common problem of data scarcity, we propose a lite transformer model that
can be trained on a reasonable amount of data, which is the case of most HTR
public datasets, without the need for external data. Second, it can learn the
reading order at page-level thanks to a curriculum learning strategy, allowing
it to avoid line segmentation errors, exploit a larger context and reduce the
need for costly segmentation annotations. Third, it can be easily adapted to
other scripts by applying a simple transfer-learning process using only
page-level labeled images. Extensive experiments on different datasets with
different scripts (French, English, Spanish, and Arabic) show the effectiveness
of the proposed model.
| no_new_dataset | 0.951997 |
2303.13932 | Qinglin Zhang | Qinglin Zhang, Chong Deng, Jiaqing Liu, Hai Yu, Qian Chen, Wen Wang,
Zhijie Yan, Jinglin Liu, Yi Ren, Zhou Zhao | Overview of the ICASSP 2023 General Meeting Understanding and Generation
Challenge (MUG) | Paper accepted to the 2023 IEEE International Conference on
Acoustics, Speech, and Signal Processing (ICASSP 2023), Rhodes, Greece | null | null | null | cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | ICASSP2023 General Meeting Understanding and Generation Challenge (MUG)
focuses on prompting a wide range of spoken language processing (SLP) research
on meeting transcripts, as SLP applications are critical to improve users'
efficiency in grasping important information in meetings. MUG includes five
tracks, including topic segmentation, topic-level and session-level extractive
summarization, topic title generation, keyphrase extraction, and action item
detection. To facilitate MUG, we construct and release a large-scale meeting
dataset, the AliMeeting4MUG Corpus.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 11:42:19 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zhang",
"Qinglin",
""
],
[
"Deng",
"Chong",
""
],
[
"Liu",
"Jiaqing",
""
],
[
"Yu",
"Hai",
""
],
[
"Chen",
"Qian",
""
],
[
"Wang",
"Wen",
""
],
[
"Yan",
"Zhijie",
""
],
[
"Liu",
"Jinglin",
""
],
[
"Ren",
"Yi",
""
],
[
"Zhao",
"Zhou",
""
]
] | TITLE: Overview of the ICASSP 2023 General Meeting Understanding and Generation
Challenge (MUG)
ABSTRACT: ICASSP2023 General Meeting Understanding and Generation Challenge (MUG)
focuses on prompting a wide range of spoken language processing (SLP) research
on meeting transcripts, as SLP applications are critical to improve users'
efficiency in grasping important information in meetings. MUG includes five
tracks, including topic segmentation, topic-level and session-level extractive
summarization, topic title generation, keyphrase extraction, and action item
detection. To facilitate MUG, we construct and release a large-scale meeting
dataset, the AliMeeting4MUG Corpus.
| new_dataset | 0.952309 |
2303.13955 | Yichen Yang | Kun He, Xin Liu, Yichen Yang, Zhou Qin, Weigao Wen, Hui Xue, John E.
Hopcroft | PIAT: Parameter Interpolation based Adversarial Training for Image
Classification | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial training has been demonstrated to be the most effective approach
to defend against adversarial attacks. However, existing adversarial training
methods show apparent oscillations and overfitting issue in the training
process, degrading the defense efficacy. In this work, we propose a novel
framework, termed Parameter Interpolation based Adversarial Training (PIAT),
that makes full use of the historical information during training.
Specifically, at the end of each epoch, PIAT tunes the model parameters as the
interpolation of the parameters of the previous and current epochs. Besides, we
suggest to use the Normalized Mean Square Error (NMSE) to further improve the
robustness by aligning the clean and adversarial examples. Compared with other
regularization methods, NMSE focuses more on the relative magnitude of the
logits rather than the absolute magnitude. Extensive experiments on several
benchmark datasets and various networks show that our method could prominently
improve the model robustness and reduce the generalization error. Moreover, our
framework is general and could further boost the robust accuracy when combined
with other adversarial training methods.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 12:22:34 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"He",
"Kun",
""
],
[
"Liu",
"Xin",
""
],
[
"Yang",
"Yichen",
""
],
[
"Qin",
"Zhou",
""
],
[
"Wen",
"Weigao",
""
],
[
"Xue",
"Hui",
""
],
[
"Hopcroft",
"John E.",
""
]
] | TITLE: PIAT: Parameter Interpolation based Adversarial Training for Image
Classification
ABSTRACT: Adversarial training has been demonstrated to be the most effective approach
to defend against adversarial attacks. However, existing adversarial training
methods show apparent oscillations and overfitting issue in the training
process, degrading the defense efficacy. In this work, we propose a novel
framework, termed Parameter Interpolation based Adversarial Training (PIAT),
that makes full use of the historical information during training.
Specifically, at the end of each epoch, PIAT tunes the model parameters as the
interpolation of the parameters of the previous and current epochs. Besides, we
suggest to use the Normalized Mean Square Error (NMSE) to further improve the
robustness by aligning the clean and adversarial examples. Compared with other
regularization methods, NMSE focuses more on the relative magnitude of the
logits rather than the absolute magnitude. Extensive experiments on several
benchmark datasets and various networks show that our method could prominently
improve the model robustness and reduce the generalization error. Moreover, our
framework is general and could further boost the robust accuracy when combined
with other adversarial training methods.
| no_new_dataset | 0.94743 |
2303.13964 | Samuel Vaiter | Hashem Ghanem (IMB), Samuel Vaiter (CNRS, JAD), Nicolas Keriven (CNRS,
IRISA) | Gradient scarcity with Bilevel Optimization for Graph Learning | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A common issue in graph learning under the semi-supervised setting is
referred to as gradient scarcity. That is, learning graphs by minimizing a loss
on a subset of nodes causes edges between unlabelled nodes that are far from
labelled ones to receive zero gradients. The phenomenon was first described
when optimizing the graph and the weights of a Graph Neural Network (GCN) with
a joint optimization algorithm. In this work, we give a precise mathematical
characterization of this phenomenon, and prove that it also emerges in bilevel
optimization, where additional dependency exists between the parameters of the
problem. While for GCNs gradient scarcity occurs due to their finite receptive
field, we show that it also occurs with the Laplacian regularization model, in
the sense that gradients amplitude decreases exponentially with distance to
labelled nodes. To alleviate this issue, we study several solutions: we propose
to resort to latent graph learning using a Graph-to-Graph model (G2G), graph
regularization to impose a prior structure on the graph, or optimizing on a
larger graph than the original one with a reduced diameter. Our experiments on
synthetic and real datasets validate our analysis and prove the efficiency of
the proposed solutions.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 12:37:43 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ghanem",
"Hashem",
"",
"IMB"
],
[
"Vaiter",
"Samuel",
"",
"CNRS, JAD"
],
[
"Keriven",
"Nicolas",
"",
"CNRS,\n IRISA"
]
] | TITLE: Gradient scarcity with Bilevel Optimization for Graph Learning
ABSTRACT: A common issue in graph learning under the semi-supervised setting is
referred to as gradient scarcity. That is, learning graphs by minimizing a loss
on a subset of nodes causes edges between unlabelled nodes that are far from
labelled ones to receive zero gradients. The phenomenon was first described
when optimizing the graph and the weights of a Graph Neural Network (GCN) with
a joint optimization algorithm. In this work, we give a precise mathematical
characterization of this phenomenon, and prove that it also emerges in bilevel
optimization, where additional dependency exists between the parameters of the
problem. While for GCNs gradient scarcity occurs due to their finite receptive
field, we show that it also occurs with the Laplacian regularization model, in
the sense that gradients amplitude decreases exponentially with distance to
labelled nodes. To alleviate this issue, we study several solutions: we propose
to resort to latent graph learning using a Graph-to-Graph model (G2G), graph
regularization to impose a prior structure on the graph, or optimizing on a
larger graph than the original one with a reduced diameter. Our experiments on
synthetic and real datasets validate our analysis and prove the efficiency of
the proposed solutions.
| no_new_dataset | 0.948442 |
2303.13971 | Yicheng Luo | Yicheng Luo, Zhengyao Jiang, Samuel Cohen, Edward Grefenstette, Marc
Peter Deisenroth | Optimal Transport for Offline Imitation Learning | Published in ICLR 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | With the advent of large datasets, offline reinforcement learning (RL) is a
promising framework for learning good decision-making policies without the need
to interact with the real environment. However, offline RL requires the dataset
to be reward-annotated, which presents practical challenges when reward
engineering is difficult or when obtaining reward annotations is
labor-intensive. In this paper, we introduce Optimal Transport Reward labeling
(OTR), an algorithm that assigns rewards to offline trajectories, with a few
high-quality demonstrations. OTR's key idea is to use optimal transport to
compute an optimal alignment between an unlabeled trajectory in the dataset and
an expert demonstration to obtain a similarity measure that can be interpreted
as a reward, which can then be used by an offline RL algorithm to learn the
policy. OTR is easy to implement and computationally efficient. On D4RL
benchmarks, we show that OTR with a single demonstration can consistently match
the performance of offline RL with ground-truth rewards.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 12:45:42 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Luo",
"Yicheng",
""
],
[
"Jiang",
"Zhengyao",
""
],
[
"Cohen",
"Samuel",
""
],
[
"Grefenstette",
"Edward",
""
],
[
"Deisenroth",
"Marc Peter",
""
]
] | TITLE: Optimal Transport for Offline Imitation Learning
ABSTRACT: With the advent of large datasets, offline reinforcement learning (RL) is a
promising framework for learning good decision-making policies without the need
to interact with the real environment. However, offline RL requires the dataset
to be reward-annotated, which presents practical challenges when reward
engineering is difficult or when obtaining reward annotations is
labor-intensive. In this paper, we introduce Optimal Transport Reward labeling
(OTR), an algorithm that assigns rewards to offline trajectories, with a few
high-quality demonstrations. OTR's key idea is to use optimal transport to
compute an optimal alignment between an unlabeled trajectory in the dataset and
an expert demonstration to obtain a similarity measure that can be interpreted
as a reward, which can then be used by an offline RL algorithm to learn the
policy. OTR is easy to implement and computationally efficient. On D4RL
benchmarks, we show that OTR with a single demonstration can consistently match
the performance of offline RL with ground-truth rewards.
| no_new_dataset | 0.942135 |
2303.13989 | Jonas Becker | Jonas Becker and Jan Philip Wahle and Terry Ruas and Bela Gipp | Paraphrase Detection: Human vs. Machine Content | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-sa/4.0/ | The growing prominence of large language models, such as GPT-4 and ChatGPT,
has led to increased concerns over academic integrity due to the potential for
machine-generated content and paraphrasing. Although studies have explored the
detection of human- and machine-paraphrased content, the comparison between
these types of content remains underexplored. In this paper, we conduct a
comprehensive analysis of various datasets commonly employed for paraphrase
detection tasks and evaluate an array of detection methods. Our findings
highlight the strengths and limitations of different detection methods in terms
of performance on individual datasets, revealing a lack of suitable
machine-generated datasets that can be aligned with human expectations. Our
main finding is that human-authored paraphrases exceed machine-generated ones
in terms of difficulty, diversity, and similarity implying that automatically
generated texts are not yet on par with human-level performance. Transformers
emerged as the most effective method across datasets with TF-IDF excelling on
semantically diverse corpora. Additionally, we identify four datasets as the
most diverse and challenging for paraphrase detection.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 13:25:46 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Becker",
"Jonas",
""
],
[
"Wahle",
"Jan Philip",
""
],
[
"Ruas",
"Terry",
""
],
[
"Gipp",
"Bela",
""
]
] | TITLE: Paraphrase Detection: Human vs. Machine Content
ABSTRACT: The growing prominence of large language models, such as GPT-4 and ChatGPT,
has led to increased concerns over academic integrity due to the potential for
machine-generated content and paraphrasing. Although studies have explored the
detection of human- and machine-paraphrased content, the comparison between
these types of content remains underexplored. In this paper, we conduct a
comprehensive analysis of various datasets commonly employed for paraphrase
detection tasks and evaluate an array of detection methods. Our findings
highlight the strengths and limitations of different detection methods in terms
of performance on individual datasets, revealing a lack of suitable
machine-generated datasets that can be aligned with human expectations. Our
main finding is that human-authored paraphrases exceed machine-generated ones
in terms of difficulty, diversity, and similarity implying that automatically
generated texts are not yet on par with human-level performance. Transformers
emerged as the most effective method across datasets with TF-IDF excelling on
semantically diverse corpora. Additionally, we identify four datasets as the
most diverse and challenging for paraphrase detection.
| no_new_dataset | 0.944893 |
2303.13995 | Seong Tae Kim | Yong Hyun Ahn, Gyeong-Moon Park, Seong Tae Kim | LINe: Out-of-Distribution Detection by Leveraging Important Neurons | IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | It is important to quantify the uncertainty of input samples, especially in
mission-critical domains such as autonomous driving and healthcare, where
failure predictions on out-of-distribution (OOD) data are likely to cause big
problems. OOD detection problem fundamentally begins in that the model cannot
express what it is not aware of. Post-hoc OOD detection approaches are widely
explored because they do not require an additional re-training process which
might degrade the model's performance and increase the training cost. In this
study, from the perspective of neurons in the deep layer of the model
representing high-level features, we introduce a new aspect for analyzing the
difference in model outputs between in-distribution data and OOD data. We
propose a novel method, Leveraging Important Neurons (LINe), for post-hoc Out
of distribution detection.
Shapley value-based pruning reduces the effects of noisy outputs by selecting
only high-contribution neurons for predicting specific classes of input data
and masking the rest. Activation clipping fixes all values above a certain
threshold into the same value, allowing LINe to treat all the class-specific
features equally and just consider the difference between the number of
activated feature differences between in-distribution and OOD data.
Comprehensive experiments verify the effectiveness of the proposed method by
outperforming state-of-the-art post-hoc OOD detection methods on CIFAR-10,
CIFAR-100, and ImageNet datasets.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 13:49:05 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ahn",
"Yong Hyun",
""
],
[
"Park",
"Gyeong-Moon",
""
],
[
"Kim",
"Seong Tae",
""
]
] | TITLE: LINe: Out-of-Distribution Detection by Leveraging Important Neurons
ABSTRACT: It is important to quantify the uncertainty of input samples, especially in
mission-critical domains such as autonomous driving and healthcare, where
failure predictions on out-of-distribution (OOD) data are likely to cause big
problems. OOD detection problem fundamentally begins in that the model cannot
express what it is not aware of. Post-hoc OOD detection approaches are widely
explored because they do not require an additional re-training process which
might degrade the model's performance and increase the training cost. In this
study, from the perspective of neurons in the deep layer of the model
representing high-level features, we introduce a new aspect for analyzing the
difference in model outputs between in-distribution data and OOD data. We
propose a novel method, Leveraging Important Neurons (LINe), for post-hoc Out
of distribution detection.
Shapley value-based pruning reduces the effects of noisy outputs by selecting
only high-contribution neurons for predicting specific classes of input data
and masking the rest. Activation clipping fixes all values above a certain
threshold into the same value, allowing LINe to treat all the class-specific
features equally and just consider the difference between the number of
activated feature differences between in-distribution and OOD data.
Comprehensive experiments verify the effectiveness of the proposed method by
outperforming state-of-the-art post-hoc OOD detection methods on CIFAR-10,
CIFAR-100, and ImageNet datasets.
| no_new_dataset | 0.949529 |
2303.14004 | Ramachandra Raghavendra Prof. | Raghavendra Ramachandra, Sushma Venkatesh, Gaurav Jaswal, Guoqiang Li | Vulnerability of Face Morphing Attacks: A Case Study on Lookalike and
Identical Twins | Accepted in IWBF 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Face morphing attacks have emerged as a potential threat, particularly in
automatic border control scenarios. Morphing attacks permit more than one
individual to use travel documents that can be used to cross borders using
automatic border control gates. The potential for morphing attacks depends on
the selection of data subjects (accomplice and malicious actors). This work
investigates lookalike and identical twins as the source of face morphing
generation. We present a systematic study on benchmarking the vulnerability of
Face Recognition Systems (FRS) to lookalike and identical twin morphing images.
Therefore, we constructed new face morphing datasets using 16 pairs of
identical twin and lookalike data subjects. Morphing images from lookalike and
identical twins are generated using a landmark-based method. Extensive
experiments are carried out to benchmark the attack potential of lookalike and
identical twins. Furthermore, experiments are designed to provide insights into
the impact of vulnerability with normal face morphing compared with lookalike
and identical twin face morphing.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 13:59:48 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Ramachandra",
"Raghavendra",
""
],
[
"Venkatesh",
"Sushma",
""
],
[
"Jaswal",
"Gaurav",
""
],
[
"Li",
"Guoqiang",
""
]
] | TITLE: Vulnerability of Face Morphing Attacks: A Case Study on Lookalike and
Identical Twins
ABSTRACT: Face morphing attacks have emerged as a potential threat, particularly in
automatic border control scenarios. Morphing attacks permit more than one
individual to use travel documents that can be used to cross borders using
automatic border control gates. The potential for morphing attacks depends on
the selection of data subjects (accomplice and malicious actors). This work
investigates lookalike and identical twins as the source of face morphing
generation. We present a systematic study on benchmarking the vulnerability of
Face Recognition Systems (FRS) to lookalike and identical twin morphing images.
Therefore, we constructed new face morphing datasets using 16 pairs of
identical twin and lookalike data subjects. Morphing images from lookalike and
identical twins are generated using a landmark-based method. Extensive
experiments are carried out to benchmark the attack potential of lookalike and
identical twins. Furthermore, experiments are designed to provide insights into
the impact of vulnerability with normal face morphing compared with lookalike
and identical twin face morphing.
| new_dataset | 0.953188 |
2303.14005 | Chi Xie | Chi Xie, Fangao Zeng, Yue Hu, Shuang Liang and Yichen Wei | Category Query Learning for Human-Object Interaction Classification | Accepted by CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Unlike most previous HOI methods that focus on learning better human-object
features, we propose a novel and complementary approach called category query
learning. Such queries are explicitly associated to interaction categories,
converted to image specific category representation via a transformer decoder,
and learnt via an auxiliary image-level classification task. This idea is
motivated by an earlier multi-label image classification method, but is for the
first time applied for the challenging human-object interaction classification
task. Our method is simple, general and effective. It is validated on three
representative HOI baselines and achieves new state-of-the-art results on two
benchmarks.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 13:59:58 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Xie",
"Chi",
""
],
[
"Zeng",
"Fangao",
""
],
[
"Hu",
"Yue",
""
],
[
"Liang",
"Shuang",
""
],
[
"Wei",
"Yichen",
""
]
] | TITLE: Category Query Learning for Human-Object Interaction Classification
ABSTRACT: Unlike most previous HOI methods that focus on learning better human-object
features, we propose a novel and complementary approach called category query
learning. Such queries are explicitly associated to interaction categories,
converted to image specific category representation via a transformer decoder,
and learnt via an auxiliary image-level classification task. This idea is
motivated by an earlier multi-label image classification method, but is for the
first time applied for the challenging human-object interaction classification
task. Our method is simple, general and effective. It is validated on three
representative HOI baselines and achieves new state-of-the-art results on two
benchmarks.
| no_new_dataset | 0.949995 |
2303.14009 | Lilas Alrahis | Lilas Alrahis, Satwik Patnaik, Muhammad Abdullah Hanif, Muhammad
Shafique, and Ozgur Sinanoglu | PoisonedGNN: Backdoor Attack on Graph Neural Networks-based Hardware
Security Systems | This manuscript is currently under review at IEEE Transactions on
Computers | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph neural networks (GNNs) have shown great success in detecting
intellectual property (IP) piracy and hardware Trojans (HTs). However, the
machine learning community has demonstrated that GNNs are susceptible to data
poisoning attacks, which result in GNNs performing abnormally on graphs with
pre-defined backdoor triggers (realized using crafted subgraphs). Thus, it is
imperative to ensure that the adoption of GNNs should not introduce security
vulnerabilities in critical security frameworks.
Existing backdoor attacks on GNNs generate random subgraphs with specific
sizes/densities to act as backdoor triggers. However, for Boolean circuits,
backdoor triggers cannot be randomized since the added structures should not
affect the functionality of a design.
We explore this threat and develop PoisonedGNN as the first backdoor attack
on GNNs in the context of hardware design. We design and inject backdoor
triggers into the register-transfer- or the gate-level representation of a
given design without affecting the functionality to evade some GNN-based
detection procedures. To demonstrate the effectiveness of PoisonedGNN, we
consider two case studies: (i) Hiding HTs and (ii) IP piracy. Our experiments
on TrustHub datasets demonstrate that PoisonedGNN can hide HTs and IP piracy
from advanced GNN-based detection platforms with an attack success rate of up
to 100%.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 14:03:55 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Alrahis",
"Lilas",
""
],
[
"Patnaik",
"Satwik",
""
],
[
"Hanif",
"Muhammad Abdullah",
""
],
[
"Shafique",
"Muhammad",
""
],
[
"Sinanoglu",
"Ozgur",
""
]
] | TITLE: PoisonedGNN: Backdoor Attack on Graph Neural Networks-based Hardware
Security Systems
ABSTRACT: Graph neural networks (GNNs) have shown great success in detecting
intellectual property (IP) piracy and hardware Trojans (HTs). However, the
machine learning community has demonstrated that GNNs are susceptible to data
poisoning attacks, which result in GNNs performing abnormally on graphs with
pre-defined backdoor triggers (realized using crafted subgraphs). Thus, it is
imperative to ensure that the adoption of GNNs should not introduce security
vulnerabilities in critical security frameworks.
Existing backdoor attacks on GNNs generate random subgraphs with specific
sizes/densities to act as backdoor triggers. However, for Boolean circuits,
backdoor triggers cannot be randomized since the added structures should not
affect the functionality of a design.
We explore this threat and develop PoisonedGNN as the first backdoor attack
on GNNs in the context of hardware design. We design and inject backdoor
triggers into the register-transfer- or the gate-level representation of a
given design without affecting the functionality to evade some GNN-based
detection procedures. To demonstrate the effectiveness of PoisonedGNN, we
consider two case studies: (i) Hiding HTs and (ii) IP piracy. Our experiments
on TrustHub datasets demonstrate that PoisonedGNN can hide HTs and IP piracy
from advanced GNN-based detection platforms with an attack success rate of up
to 100%.
| no_new_dataset | 0.939665 |
2303.14038 | Teng Wang | Teng Wang, Yixiao Ge, Feng Zheng, Ran Cheng, Ying Shan, Xiaohu Qie,
Ping Luo | Accelerating Vision-Language Pretraining with Free Language Modeling | To appear in CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The state of the arts in vision-language pretraining (VLP) achieves exemplary
performance but suffers from high training costs resulting from slow
convergence and long training time, especially on large-scale web datasets. An
essential obstacle to training efficiency lies in the entangled prediction rate
(percentage of tokens for reconstruction) and corruption rate (percentage of
corrupted tokens) in masked language modeling (MLM), that is, a proper
corruption rate is achieved at the cost of a large portion of output tokens
being excluded from prediction loss. To accelerate the convergence of VLP, we
propose a new pretraining task, namely, free language modeling (FLM), that
enables a 100% prediction rate with arbitrary corruption rates. FLM
successfully frees the prediction rate from the tie-up with the corruption rate
while allowing the corruption spans to be customized for each token to be
predicted. FLM-trained models are encouraged to learn better and faster given
the same GPU time by exploiting bidirectional contexts more flexibly. Extensive
experiments show FLM could achieve an impressive 2.5x pretraining time
reduction in comparison to the MLM-based methods, while keeping competitive
performance on both vision-language understanding and generation tasks. Code
will be public at https://github.com/TencentARC/FLM.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 14:49:22 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wang",
"Teng",
""
],
[
"Ge",
"Yixiao",
""
],
[
"Zheng",
"Feng",
""
],
[
"Cheng",
"Ran",
""
],
[
"Shan",
"Ying",
""
],
[
"Qie",
"Xiaohu",
""
],
[
"Luo",
"Ping",
""
]
] | TITLE: Accelerating Vision-Language Pretraining with Free Language Modeling
ABSTRACT: The state of the arts in vision-language pretraining (VLP) achieves exemplary
performance but suffers from high training costs resulting from slow
convergence and long training time, especially on large-scale web datasets. An
essential obstacle to training efficiency lies in the entangled prediction rate
(percentage of tokens for reconstruction) and corruption rate (percentage of
corrupted tokens) in masked language modeling (MLM), that is, a proper
corruption rate is achieved at the cost of a large portion of output tokens
being excluded from prediction loss. To accelerate the convergence of VLP, we
propose a new pretraining task, namely, free language modeling (FLM), that
enables a 100% prediction rate with arbitrary corruption rates. FLM
successfully frees the prediction rate from the tie-up with the corruption rate
while allowing the corruption spans to be customized for each token to be
predicted. FLM-trained models are encouraged to learn better and faster given
the same GPU time by exploiting bidirectional contexts more flexibly. Extensive
experiments show FLM could achieve an impressive 2.5x pretraining time
reduction in comparison to the MLM-based methods, while keeping competitive
performance on both vision-language understanding and generation tasks. Code
will be public at https://github.com/TencentARC/FLM.
| no_new_dataset | 0.949248 |
2303.14075 | Zeyong Wei | Mingqiang Wei, Qian Sun, Haoran Xie, Dong Liang, Fu Lee Wang | Search By Image: Deeply Exploring Beneficial Features for Beauty Product
Retrieval | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Searching by image is popular yet still challenging due to the extensive
interference arose from i) data variations (e.g., background, pose, visual
angle, brightness) of real-world captured images and ii) similar images in the
query dataset. This paper studies a practically meaningful problem of beauty
product retrieval (BPR) by neural networks. We broadly extract different types
of image features, and raise an intriguing question that whether these features
are beneficial to i) suppress data variations of real-world captured images,
and ii) distinguish one image from others which look very similar but are
intrinsically different beauty products in the dataset, therefore leading to an
enhanced capability of BPR. To answer it, we present a novel variable-attention
neural network to understand the combination of multiple features (termed
VM-Net) of beauty product images. Considering that there are few publicly
released training datasets for BPR, we establish a new dataset with more than
one million images classified into more than 20K categories to improve both the
generalization and anti-interference abilities of VM-Net and other methods. We
verify the performance of VM-Net and its competitors on the benchmark dataset
Perfect-500K, where VM-Net shows clear improvements over the competitors in
terms of MAP@7. The source code and dataset will be released upon publication.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 15:38:58 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wei",
"Mingqiang",
""
],
[
"Sun",
"Qian",
""
],
[
"Xie",
"Haoran",
""
],
[
"Liang",
"Dong",
""
],
[
"Wang",
"Fu Lee",
""
]
] | TITLE: Search By Image: Deeply Exploring Beneficial Features for Beauty Product
Retrieval
ABSTRACT: Searching by image is popular yet still challenging due to the extensive
interference arose from i) data variations (e.g., background, pose, visual
angle, brightness) of real-world captured images and ii) similar images in the
query dataset. This paper studies a practically meaningful problem of beauty
product retrieval (BPR) by neural networks. We broadly extract different types
of image features, and raise an intriguing question that whether these features
are beneficial to i) suppress data variations of real-world captured images,
and ii) distinguish one image from others which look very similar but are
intrinsically different beauty products in the dataset, therefore leading to an
enhanced capability of BPR. To answer it, we present a novel variable-attention
neural network to understand the combination of multiple features (termed
VM-Net) of beauty product images. Considering that there are few publicly
released training datasets for BPR, we establish a new dataset with more than
one million images classified into more than 20K categories to improve both the
generalization and anti-interference abilities of VM-Net and other methods. We
verify the performance of VM-Net and its competitors on the benchmark dataset
Perfect-500K, where VM-Net shows clear improvements over the competitors in
terms of MAP@7. The source code and dataset will be released upon publication.
| new_dataset | 0.965576 |
2303.14078 | Jisoo Jeong | Jisoo Jeong, Hong Cai, Risheek Garrepalli, Fatih Porikli | DistractFlow: Improving Optical Flow Estimation via Realistic
Distractions and Pseudo-Labeling | CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose a novel data augmentation approach, DistractFlow, for training
optical flow estimation models by introducing realistic distractions to the
input frames. Based on a mixing ratio, we combine one of the frames in the pair
with a distractor image depicting a similar domain, which allows for inducing
visual perturbations congruent with natural objects and scenes. We refer to
such pairs as distracted pairs. Our intuition is that using semantically
meaningful distractors enables the model to learn related variations and attain
robustness against challenging deviations, compared to conventional
augmentation schemes focusing only on low-level aspects and modifications. More
specifically, in addition to the supervised loss computed between the estimated
flow for the original pair and its ground-truth flow, we include a second
supervised loss defined between the distracted pair's flow and the original
pair's ground-truth flow, weighted with the same mixing ratio. Furthermore,
when unlabeled data is available, we extend our augmentation approach to
self-supervised settings through pseudo-labeling and cross-consistency
regularization. Given an original pair and its distracted version, we enforce
the estimated flow on the distracted pair to agree with the flow of the
original pair. Our approach allows increasing the number of available training
pairs significantly without requiring additional annotations. It is agnostic to
the model architecture and can be applied to training any optical flow
estimation models. Our extensive evaluations on multiple benchmarks, including
Sintel, KITTI, and SlowFlow, show that DistractFlow improves existing models
consistently, outperforming the latest state of the art.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 15:42:54 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Jeong",
"Jisoo",
""
],
[
"Cai",
"Hong",
""
],
[
"Garrepalli",
"Risheek",
""
],
[
"Porikli",
"Fatih",
""
]
] | TITLE: DistractFlow: Improving Optical Flow Estimation via Realistic
Distractions and Pseudo-Labeling
ABSTRACT: We propose a novel data augmentation approach, DistractFlow, for training
optical flow estimation models by introducing realistic distractions to the
input frames. Based on a mixing ratio, we combine one of the frames in the pair
with a distractor image depicting a similar domain, which allows for inducing
visual perturbations congruent with natural objects and scenes. We refer to
such pairs as distracted pairs. Our intuition is that using semantically
meaningful distractors enables the model to learn related variations and attain
robustness against challenging deviations, compared to conventional
augmentation schemes focusing only on low-level aspects and modifications. More
specifically, in addition to the supervised loss computed between the estimated
flow for the original pair and its ground-truth flow, we include a second
supervised loss defined between the distracted pair's flow and the original
pair's ground-truth flow, weighted with the same mixing ratio. Furthermore,
when unlabeled data is available, we extend our augmentation approach to
self-supervised settings through pseudo-labeling and cross-consistency
regularization. Given an original pair and its distracted version, we enforce
the estimated flow on the distracted pair to agree with the flow of the
original pair. Our approach allows increasing the number of available training
pairs significantly without requiring additional annotations. It is agnostic to
the model architecture and can be applied to training any optical flow
estimation models. Our extensive evaluations on multiple benchmarks, including
Sintel, KITTI, and SlowFlow, show that DistractFlow improves existing models
consistently, outperforming the latest state of the art.
| no_new_dataset | 0.948775 |
2303.14087 | Xiaohao Sun | Xiaohao Sun, Hanxiao Jiang, Manolis Savva, Angel Xuan Chang | OPDMulti: Openable Part Detection for Multiple Objects | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Openable part detection is the task of detecting the openable parts of an
object in a single-view image, and predicting corresponding motion parameters.
Prior work investigated the unrealistic setting where all input images only
contain a single openable object. We generalize this task to scenes with
multiple objects each potentially possessing openable parts, and create a
corresponding dataset based on real-world scenes. We then address this more
challenging scenario with OPDFormer: a part-aware transformer architecture. Our
experiments show that the OPDFormer architecture significantly outperforms
prior work. The more realistic multiple-object scenarios we investigated remain
challenging for all methods, indicating opportunities for future work.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 15:52:20 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Sun",
"Xiaohao",
""
],
[
"Jiang",
"Hanxiao",
""
],
[
"Savva",
"Manolis",
""
],
[
"Chang",
"Angel Xuan",
""
]
] | TITLE: OPDMulti: Openable Part Detection for Multiple Objects
ABSTRACT: Openable part detection is the task of detecting the openable parts of an
object in a single-view image, and predicting corresponding motion parameters.
Prior work investigated the unrealistic setting where all input images only
contain a single openable object. We generalize this task to scenes with
multiple objects each potentially possessing openable parts, and create a
corresponding dataset based on real-world scenes. We then address this more
challenging scenario with OPDFormer: a part-aware transformer architecture. Our
experiments show that the OPDFormer architecture significantly outperforms
prior work. The more realistic multiple-object scenarios we investigated remain
challenging for all methods, indicating opportunities for future work.
| new_dataset | 0.956268 |
2303.14089 | Yaroslav Zharov | Yaroslav Zharov, Tilo Baumbach, Vincent Heuveline | Optimizing the Procedure of CT Segmentation Labeling | Under review | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | In Computed Tomography, machine learning is often used for automated data
processing. However, increasing model complexity is accompanied by increasingly
large volume datasets, which in turn increases the cost of model training.
Unlike most work that mitigates this by advancing model architectures and
training algorithms, we consider the annotation procedure and its effect on the
model performance. We assume three main virtues of a good dataset collected for
a model training to be label quality, diversity, and completeness. We compare
the effects of those virtues on the model performance using open medical CT
datasets and conclude, that quality is more important than diversity early
during labeling; the diversity, in turn, is more important than completeness.
Based on this conclusion and additional experiments, we propose a labeling
procedure for the segmentation of tomographic images to minimize efforts spent
on labeling while maximizing the model performance.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 15:52:42 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Zharov",
"Yaroslav",
""
],
[
"Baumbach",
"Tilo",
""
],
[
"Heuveline",
"Vincent",
""
]
] | TITLE: Optimizing the Procedure of CT Segmentation Labeling
ABSTRACT: In Computed Tomography, machine learning is often used for automated data
processing. However, increasing model complexity is accompanied by increasingly
large volume datasets, which in turn increases the cost of model training.
Unlike most work that mitigates this by advancing model architectures and
training algorithms, we consider the annotation procedure and its effect on the
model performance. We assume three main virtues of a good dataset collected for
a model training to be label quality, diversity, and completeness. We compare
the effects of those virtues on the model performance using open medical CT
datasets and conclude, that quality is more important than diversity early
during labeling; the diversity, in turn, is more important than completeness.
Based on this conclusion and additional experiments, we propose a labeling
procedure for the segmentation of tomographic images to minimize efforts spent
on labeling while maximizing the model performance.
| no_new_dataset | 0.952086 |
2303.14096 | Jongheon Jeong | Jongheon Jeong, Sihyun Yu, Hankook Lee, Jinwoo Shin | Enhancing Multiple Reliability Measures via Nuisance-extended
Information Bottleneck | 25 pages; CVPR 2023 | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In practical scenarios where training data is limited, many predictive
signals in the data can be rather from some biases in data acquisition (i.e.,
less generalizable), so that one cannot prevent a model from co-adapting on
such (so-called) "shortcut" signals: this makes the model fragile in various
distribution shifts. To bypass such failure modes, we consider an adversarial
threat model under a mutual information constraint to cover a wider class of
perturbations in training. This motivates us to extend the standard information
bottleneck to additionally model the nuisance information. We propose an
autoencoder-based training to implement the objective, as well as practical
encoder designs to facilitate the proposed hybrid discriminative-generative
training concerning both convolutional- and Transformer-based architectures.
Our experimental results show that the proposed scheme improves robustness of
learned representations (remarkably without using any domain-specific
knowledge), with respect to multiple challenging reliability measures. For
example, our model could advance the state-of-the-art on a recent challenging
OBJECTS benchmark in novelty detection by $78.4\% \rightarrow 87.2\%$ in AUROC,
while simultaneously enjoying improved corruption, background and (certified)
adversarial robustness. Code is available at
https://github.com/jh-jeong/nuisance_ib.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 16:03:21 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Jeong",
"Jongheon",
""
],
[
"Yu",
"Sihyun",
""
],
[
"Lee",
"Hankook",
""
],
[
"Shin",
"Jinwoo",
""
]
] | TITLE: Enhancing Multiple Reliability Measures via Nuisance-extended
Information Bottleneck
ABSTRACT: In practical scenarios where training data is limited, many predictive
signals in the data can be rather from some biases in data acquisition (i.e.,
less generalizable), so that one cannot prevent a model from co-adapting on
such (so-called) "shortcut" signals: this makes the model fragile in various
distribution shifts. To bypass such failure modes, we consider an adversarial
threat model under a mutual information constraint to cover a wider class of
perturbations in training. This motivates us to extend the standard information
bottleneck to additionally model the nuisance information. We propose an
autoencoder-based training to implement the objective, as well as practical
encoder designs to facilitate the proposed hybrid discriminative-generative
training concerning both convolutional- and Transformer-based architectures.
Our experimental results show that the proposed scheme improves robustness of
learned representations (remarkably without using any domain-specific
knowledge), with respect to multiple challenging reliability measures. For
example, our model could advance the state-of-the-art on a recent challenging
OBJECTS benchmark in novelty detection by $78.4\% \rightarrow 87.2\%$ in AUROC,
while simultaneously enjoying improved corruption, background and (certified)
adversarial robustness. Code is available at
https://github.com/jh-jeong/nuisance_ib.
| no_new_dataset | 0.947769 |
2303.14102 | Marco Gaido | Marco Gaido | Distributed Silhouette Algorithm: Evaluating Clustering on Big Data | null | null | null | null | cs.DC cs.AI cs.DS | http://creativecommons.org/licenses/by-sa/4.0/ | In the big data era, the key feature that each algorithm needs to have is the
possibility of efficiently running in parallel in a distributed environment.
The popular Silhouette metric to evaluate the quality of a clustering,
unfortunately, does not have this property and has a quadratic computational
complexity with respect to the size of the input dataset. For this reason, its
execution has been hindered in big data scenarios, where clustering had to be
evaluated otherwise. To fill this gap, in this paper we introduce the first
algorithm that computes the Silhouette metric with linear complexity and can
easily execute in parallel in a distributed environment. Its implementation is
freely available in the Apache Spark ML library.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 16:10:43 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Gaido",
"Marco",
""
]
] | TITLE: Distributed Silhouette Algorithm: Evaluating Clustering on Big Data
ABSTRACT: In the big data era, the key feature that each algorithm needs to have is the
possibility of efficiently running in parallel in a distributed environment.
The popular Silhouette metric to evaluate the quality of a clustering,
unfortunately, does not have this property and has a quadratic computational
complexity with respect to the size of the input dataset. For this reason, its
execution has been hindered in big data scenarios, where clustering had to be
evaluated otherwise. To fill this gap, in this paper we introduce the first
algorithm that computes the Silhouette metric with linear complexity and can
easily execute in parallel in a distributed environment. Its implementation is
freely available in the Apache Spark ML library.
| no_new_dataset | 0.951684 |
2303.14123 | Wentao Chen | Wentao Chen, Chenyang Si, Zhang Zhang, Liang Wang, Zilei Wang, Tieniu
Tan | Semantic Prompt for Few-Shot Image Recognition | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot learning is a challenging problem since only a few examples are
provided to recognize a new class. Several recent studies exploit additional
semantic information, e.g. text embeddings of class names, to address the issue
of rare samples through combining semantic prototypes with visual prototypes.
However, these methods still suffer from the spurious visual features learned
from the rare support samples, resulting in limited benefits. In this paper, we
propose a novel Semantic Prompt (SP) approach for few-shot learning. Instead of
the naive exploitation of semantic information for remedying classifiers, we
explore leveraging semantic information as prompts to tune the visual feature
extraction network adaptively. Specifically, we design two complementary
mechanisms to insert semantic prompts into the feature extractor: one is to
enable the interaction between semantic prompts and patch embeddings along the
spatial dimension via self-attention, another is to supplement visual features
with the transformed semantic prompts along the channel dimension. By combining
these two mechanisms, the feature extractor presents a better ability to attend
to the class-specific features and obtains more generalized image
representations with merely a few support samples. Through extensive
experiments on four datasets, the proposed approach achieves promising results,
improving the 1-shot learning accuracy by 3.67% on average.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 16:32:19 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Chen",
"Wentao",
""
],
[
"Si",
"Chenyang",
""
],
[
"Zhang",
"Zhang",
""
],
[
"Wang",
"Liang",
""
],
[
"Wang",
"Zilei",
""
],
[
"Tan",
"Tieniu",
""
]
] | TITLE: Semantic Prompt for Few-Shot Image Recognition
ABSTRACT: Few-shot learning is a challenging problem since only a few examples are
provided to recognize a new class. Several recent studies exploit additional
semantic information, e.g. text embeddings of class names, to address the issue
of rare samples through combining semantic prototypes with visual prototypes.
However, these methods still suffer from the spurious visual features learned
from the rare support samples, resulting in limited benefits. In this paper, we
propose a novel Semantic Prompt (SP) approach for few-shot learning. Instead of
the naive exploitation of semantic information for remedying classifiers, we
explore leveraging semantic information as prompts to tune the visual feature
extraction network adaptively. Specifically, we design two complementary
mechanisms to insert semantic prompts into the feature extractor: one is to
enable the interaction between semantic prompts and patch embeddings along the
spatial dimension via self-attention, another is to supplement visual features
with the transformed semantic prompts along the channel dimension. By combining
these two mechanisms, the feature extractor presents a better ability to attend
to the class-specific features and obtains more generalized image
representations with merely a few support samples. Through extensive
experiments on four datasets, the proposed approach achieves promising results,
improving the 1-shot learning accuracy by 3.67% on average.
| no_new_dataset | 0.948775 |
2303.14124 | Bo He | Bo He, Xitong Yang, Hanyu Wang, Zuxuan Wu, Hao Chen, Shuaiyi Huang,
Yixuan Ren, Ser-Nam Lim, Abhinav Shrivastava | Towards Scalable Neural Representation for Diverse Videos | Accepted at CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Implicit neural representations (INR) have gained increasing attention in
representing 3D scenes and images, and have been recently applied to encode
videos (e.g., NeRV, E-NeRV). While achieving promising results, existing
INR-based methods are limited to encoding a handful of short videos (e.g.,
seven 5-second videos in the UVG dataset) with redundant visual content,
leading to a model design that fits individual video frames independently and
is not efficiently scalable to a large number of diverse videos. This paper
focuses on developing neural representations for a more practical setup --
encoding long and/or a large number of videos with diverse visual content. We
first show that instead of dividing videos into small subsets and encoding them
with separate models, encoding long and diverse videos jointly with a unified
model achieves better compression results. Based on this observation, we
propose D-NeRV, a novel neural representation framework designed to encode
diverse videos by (i) decoupling clip-specific visual content from motion
information, (ii) introducing temporal reasoning into the implicit neural
network, and (iii) employing the task-oriented flow as intermediate output to
reduce spatial redundancies. Our new model largely surpasses NeRV and
traditional video compression techniques on UCF101 and UVG datasets on the
video compression task. Moreover, when used as an efficient data-loader, D-NeRV
achieves 3%-10% higher accuracy than NeRV on action recognition tasks on the
UCF101 dataset under the same compression ratios.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 16:32:19 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"He",
"Bo",
""
],
[
"Yang",
"Xitong",
""
],
[
"Wang",
"Hanyu",
""
],
[
"Wu",
"Zuxuan",
""
],
[
"Chen",
"Hao",
""
],
[
"Huang",
"Shuaiyi",
""
],
[
"Ren",
"Yixuan",
""
],
[
"Lim",
"Ser-Nam",
""
],
[
"Shrivastava",
"Abhinav",
""
]
] | TITLE: Towards Scalable Neural Representation for Diverse Videos
ABSTRACT: Implicit neural representations (INR) have gained increasing attention in
representing 3D scenes and images, and have been recently applied to encode
videos (e.g., NeRV, E-NeRV). While achieving promising results, existing
INR-based methods are limited to encoding a handful of short videos (e.g.,
seven 5-second videos in the UVG dataset) with redundant visual content,
leading to a model design that fits individual video frames independently and
is not efficiently scalable to a large number of diverse videos. This paper
focuses on developing neural representations for a more practical setup --
encoding long and/or a large number of videos with diverse visual content. We
first show that instead of dividing videos into small subsets and encoding them
with separate models, encoding long and diverse videos jointly with a unified
model achieves better compression results. Based on this observation, we
propose D-NeRV, a novel neural representation framework designed to encode
diverse videos by (i) decoupling clip-specific visual content from motion
information, (ii) introducing temporal reasoning into the implicit neural
network, and (iii) employing the task-oriented flow as intermediate output to
reduce spatial redundancies. Our new model largely surpasses NeRV and
traditional video compression techniques on UCF101 and UVG datasets on the
video compression task. Moreover, when used as an efficient data-loader, D-NeRV
achieves 3%-10% higher accuracy than NeRV on action recognition tasks on the
UCF101 dataset under the same compression ratios.
| no_new_dataset | 0.947284 |
2303.14126 | Jordan J. Bird | Jordan J. Bird, Ahmad Lotfi | CIFAKE: Image Classification and Explainable Identification of
AI-Generated Synthetic Images | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recent technological advances in synthetic data have enabled the generation
of images with such high quality that human beings cannot tell the difference
between real-life photographs and Artificial Intelligence (AI) generated
images. Given the critical necessity of data reliability and authentication,
this article proposes to enhance our ability to recognise AI-generated images
through computer vision. Initially, a synthetic dataset is generated that
mirrors the ten classes of the already available CIFAR-10 dataset with latent
diffusion which provides a contrasting set of images for comparison to real
photographs. The model is capable of generating complex visual attributes, such
as photorealistic reflections in water. The two sets of data present as a
binary classification problem with regard to whether the photograph is real or
generated by AI. This study then proposes the use of a Convolutional Neural
Network (CNN) to classify the images into two categories; Real or Fake.
Following hyperparameter tuning and the training of 36 individual network
topologies, the optimal approach could correctly classify the images with
92.98% accuracy. Finally, this study implements explainable AI via Gradient
Class Activation Mapping to explore which features within the images are useful
for classification. Interpretation reveals interesting concepts within the
image, in particular, noting that the actual entity itself does not hold useful
information for classification; instead, the model focuses on small visual
imperfections in the background of the images. The complete dataset engineered
for this study, referred to as the CIFAKE dataset, is made publicly available
to the research community for future work.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 16:33:06 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Bird",
"Jordan J.",
""
],
[
"Lotfi",
"Ahmad",
""
]
] | TITLE: CIFAKE: Image Classification and Explainable Identification of
AI-Generated Synthetic Images
ABSTRACT: Recent technological advances in synthetic data have enabled the generation
of images with such high quality that human beings cannot tell the difference
between real-life photographs and Artificial Intelligence (AI) generated
images. Given the critical necessity of data reliability and authentication,
this article proposes to enhance our ability to recognise AI-generated images
through computer vision. Initially, a synthetic dataset is generated that
mirrors the ten classes of the already available CIFAR-10 dataset with latent
diffusion which provides a contrasting set of images for comparison to real
photographs. The model is capable of generating complex visual attributes, such
as photorealistic reflections in water. The two sets of data present as a
binary classification problem with regard to whether the photograph is real or
generated by AI. This study then proposes the use of a Convolutional Neural
Network (CNN) to classify the images into two categories; Real or Fake.
Following hyperparameter tuning and the training of 36 individual network
topologies, the optimal approach could correctly classify the images with
92.98% accuracy. Finally, this study implements explainable AI via Gradient
Class Activation Mapping to explore which features within the images are useful
for classification. Interpretation reveals interesting concepts within the
image, in particular, noting that the actual entity itself does not hold useful
information for classification; instead, the model focuses on small visual
imperfections in the background of the images. The complete dataset engineered
for this study, referred to as the CIFAKE dataset, is made publicly available
to the research community for future work.
| new_dataset | 0.973393 |
2303.14158 | Bowen Wen | Bowen Wen, Jonathan Tremblay, Valts Blukis, Stephen Tyree, Thomas
Muller, Alex Evans, Dieter Fox, Jan Kautz, Stan Birchfield | BundleSDF: Neural 6-DoF Tracking and 3D Reconstruction of Unknown
Objects | CVPR 2023 | null | null | null | cs.CV cs.AI cs.GR cs.RO | http://creativecommons.org/licenses/by/4.0/ | We present a near real-time method for 6-DoF tracking of an unknown object
from a monocular RGBD video sequence, while simultaneously performing neural 3D
reconstruction of the object. Our method works for arbitrary rigid objects,
even when visual texture is largely absent. The object is assumed to be
segmented in the first frame only. No additional information is required, and
no assumption is made about the interaction agent. Key to our method is a
Neural Object Field that is learned concurrently with a pose graph optimization
process in order to robustly accumulate information into a consistent 3D
representation capturing both geometry and appearance. A dynamic pool of posed
memory frames is automatically maintained to facilitate communication between
these threads. Our approach handles challenging sequences with large pose
changes, partial and full occlusion, untextured surfaces, and specular
highlights. We show results on HO3D, YCBInEOAT, and BEHAVE datasets,
demonstrating that our method significantly outperforms existing approaches.
Project page: https://bundlesdf.github.io
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 17:13:49 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wen",
"Bowen",
""
],
[
"Tremblay",
"Jonathan",
""
],
[
"Blukis",
"Valts",
""
],
[
"Tyree",
"Stephen",
""
],
[
"Muller",
"Thomas",
""
],
[
"Evans",
"Alex",
""
],
[
"Fox",
"Dieter",
""
],
[
"Kautz",
"Jan",
""
],
[
"Birchfield",
"Stan",
""
]
] | TITLE: BundleSDF: Neural 6-DoF Tracking and 3D Reconstruction of Unknown
Objects
ABSTRACT: We present a near real-time method for 6-DoF tracking of an unknown object
from a monocular RGBD video sequence, while simultaneously performing neural 3D
reconstruction of the object. Our method works for arbitrary rigid objects,
even when visual texture is largely absent. The object is assumed to be
segmented in the first frame only. No additional information is required, and
no assumption is made about the interaction agent. Key to our method is a
Neural Object Field that is learned concurrently with a pose graph optimization
process in order to robustly accumulate information into a consistent 3D
representation capturing both geometry and appearance. A dynamic pool of posed
memory frames is automatically maintained to facilitate communication between
these threads. Our approach handles challenging sequences with large pose
changes, partial and full occlusion, untextured surfaces, and specular
highlights. We show results on HO3D, YCBInEOAT, and BEHAVE datasets,
demonstrating that our method significantly outperforms existing approaches.
Project page: https://bundlesdf.github.io
| no_new_dataset | 0.947962 |
2303.14162 | Mehrdad Morsali | Mehrdad Morsali, Mahmoud Nazzal, Abdallah Khreishah, Shaahin Angizi | IMA-GNN: In-Memory Acceleration of Centralized and Decentralized Graph
Neural Networks at the Edge | 6 pages, 8 Figures, 2 Tables | null | null | null | cs.AR | http://creativecommons.org/licenses/by/4.0/ | In this paper, we propose IMA-GNN as an In-Memory Accelerator for centralized
and decentralized Graph Neural Network inference, explore its potential in both
settings and provide a guideline for the community targeting flexible and
efficient edge computation. Leveraging IMA-GNN, we first model the computation
and communication latencies of edge devices. We then present practical case
studies on GNN-based taxi demand and supply prediction and also adopt four
large graph datasets to quantitatively compare and analyze centralized and
decentralized settings. Our cross-layer simulation results demonstrate that on
average, IMA-GNN in the centralized setting can obtain ~790x communication
speed-up compared to the decentralized GNN setting. However, the decentralized
setting performs computation ~1400x faster while reducing the power consumption
per device. This further underlines the need for a hybrid semi-decentralized
GNN approach.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 17:15:43 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Morsali",
"Mehrdad",
""
],
[
"Nazzal",
"Mahmoud",
""
],
[
"Khreishah",
"Abdallah",
""
],
[
"Angizi",
"Shaahin",
""
]
] | TITLE: IMA-GNN: In-Memory Acceleration of Centralized and Decentralized Graph
Neural Networks at the Edge
ABSTRACT: In this paper, we propose IMA-GNN as an In-Memory Accelerator for centralized
and decentralized Graph Neural Network inference, explore its potential in both
settings and provide a guideline for the community targeting flexible and
efficient edge computation. Leveraging IMA-GNN, we first model the computation
and communication latencies of edge devices. We then present practical case
studies on GNN-based taxi demand and supply prediction and also adopt four
large graph datasets to quantitatively compare and analyze centralized and
decentralized settings. Our cross-layer simulation results demonstrate that on
average, IMA-GNN in the centralized setting can obtain ~790x communication
speed-up compared to the decentralized GNN setting. However, the decentralized
setting performs computation ~1400x faster while reducing the power consumption
per device. This further underlines the need for a hybrid semi-decentralized
GNN approach.
| no_new_dataset | 0.947137 |
2303.14191 | Xiaoyang Wu | Xiaoyang Wu, Xin Wen, Xihui Liu, Hengshuang Zhao | Masked Scene Contrast: A Scalable Framework for Unsupervised 3D
Representation Learning | Accepted by CVPR 2023, available at
https://github.com/Pointcept/Pointcept | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As a pioneering work, PointContrast conducts unsupervised 3D representation
learning via leveraging contrastive learning over raw RGB-D frames and proves
its effectiveness on various downstream tasks. However, the trend of
large-scale unsupervised learning in 3D has yet to emerge due to two stumbling
blocks: the inefficiency of matching RGB-D frames as contrastive views and the
annoying mode collapse phenomenon mentioned in previous works. Turning the two
stumbling blocks into empirical stepping stones, we first propose an efficient
and effective contrastive learning framework, which generates contrastive views
directly on scene-level point clouds by a well-curated data augmentation
pipeline and a practical view mixing strategy. Second, we introduce
reconstructive learning on the contrastive learning framework with an exquisite
design of contrastive cross masks, which targets the reconstruction of point
color and surfel normal. Our Masked Scene Contrast (MSC) framework is capable
of extracting comprehensive 3D representations more efficiently and
effectively. It accelerates the pre-training procedure by at least 3x and still
achieves an uncompromised performance compared with previous work. Besides, MSC
also enables large-scale 3D pre-training across multiple datasets, which
further boosts the performance and achieves state-of-the-art fine-tuning
results on several downstream tasks, e.g., 75.5% mIoU on ScanNet semantic
segmentation validation set.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 17:59:58 GMT"
}
] | 2023-03-27T00:00:00 | [
[
"Wu",
"Xiaoyang",
""
],
[
"Wen",
"Xin",
""
],
[
"Liu",
"Xihui",
""
],
[
"Zhao",
"Hengshuang",
""
]
] | TITLE: Masked Scene Contrast: A Scalable Framework for Unsupervised 3D
Representation Learning
ABSTRACT: As a pioneering work, PointContrast conducts unsupervised 3D representation
learning via leveraging contrastive learning over raw RGB-D frames and proves
its effectiveness on various downstream tasks. However, the trend of
large-scale unsupervised learning in 3D has yet to emerge due to two stumbling
blocks: the inefficiency of matching RGB-D frames as contrastive views and the
annoying mode collapse phenomenon mentioned in previous works. Turning the two
stumbling blocks into empirical stepping stones, we first propose an efficient
and effective contrastive learning framework, which generates contrastive views
directly on scene-level point clouds by a well-curated data augmentation
pipeline and a practical view mixing strategy. Second, we introduce
reconstructive learning on the contrastive learning framework with an exquisite
design of contrastive cross masks, which targets the reconstruction of point
color and surfel normal. Our Masked Scene Contrast (MSC) framework is capable
of extracting comprehensive 3D representations more efficiently and
effectively. It accelerates the pre-training procedure by at least 3x and still
achieves an uncompromised performance compared with previous work. Besides, MSC
also enables large-scale 3D pre-training across multiple datasets, which
further boosts the performance and achieves state-of-the-art fine-tuning
results on several downstream tasks, e.g., 75.5% mIoU on ScanNet semantic
segmentation validation set.
| no_new_dataset | 0.947186 |
1908.04531 | Leon Derczynski | Gudbjartur Ingi Sigurbergsson, Leon Derczynski | Offensive Language and Hate Speech Detection for Danish | Proceedings of the Twelfth Language Resources and Evaluation
Conference | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | The presence of offensive language on social media platforms and the
implications this poses is becoming a major concern in modern society. Given
the enormous amount of content created every day, automatic methods are
required to detect and deal with this type of content. Until now, most of the
research has focused on solving the problem for the English language, while the
problem is multilingual.
We construct a Danish dataset containing user-generated comments from
\textit{Reddit} and \textit{Facebook}. It contains user generated comments from
various social media platforms, and to our knowledge, it is the first of its
kind. Our dataset is annotated to capture various types and target of offensive
language. We develop four automatic classification systems, each designed to
work for both the English and the Danish language. In the detection of
offensive language in English, the best performing system achieves a macro
averaged F1-score of $0.74$, and the best performing system for Danish achieves
a macro averaged F1-score of $0.70$. In the detection of whether or not an
offensive post is targeted, the best performing system for English achieves a
macro averaged F1-score of $0.62$, while the best performing system for Danish
achieves a macro averaged F1-score of $0.73$. Finally, in the detection of the
target type in a targeted offensive post, the best performing system for
English achieves a macro averaged F1-score of $0.56$, and the best performing
system for Danish achieves a macro averaged F1-score of $0.63$.
Our work for both the English and the Danish language captures the type and
targets of offensive language, and present automatic methods for detecting
different kinds of offensive language such as hate speech and cyberbullying.
| [
{
"version": "v1",
"created": "Tue, 13 Aug 2019 08:29:48 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 04:24:09 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Sigurbergsson",
"Gudbjartur Ingi",
""
],
[
"Derczynski",
"Leon",
""
]
] | TITLE: Offensive Language and Hate Speech Detection for Danish
ABSTRACT: The presence of offensive language on social media platforms and the
implications this poses is becoming a major concern in modern society. Given
the enormous amount of content created every day, automatic methods are
required to detect and deal with this type of content. Until now, most of the
research has focused on solving the problem for the English language, while the
problem is multilingual.
We construct a Danish dataset containing user-generated comments from
\textit{Reddit} and \textit{Facebook}. It contains user generated comments from
various social media platforms, and to our knowledge, it is the first of its
kind. Our dataset is annotated to capture various types and target of offensive
language. We develop four automatic classification systems, each designed to
work for both the English and the Danish language. In the detection of
offensive language in English, the best performing system achieves a macro
averaged F1-score of $0.74$, and the best performing system for Danish achieves
a macro averaged F1-score of $0.70$. In the detection of whether or not an
offensive post is targeted, the best performing system for English achieves a
macro averaged F1-score of $0.62$, while the best performing system for Danish
achieves a macro averaged F1-score of $0.73$. Finally, in the detection of the
target type in a targeted offensive post, the best performing system for
English achieves a macro averaged F1-score of $0.56$, and the best performing
system for Danish achieves a macro averaged F1-score of $0.63$.
Our work for both the English and the Danish language captures the type and
targets of offensive language, and present automatic methods for detecting
different kinds of offensive language such as hate speech and cyberbullying.
| new_dataset | 0.969382 |
2004.01670 | Bertie Vidgen Dr | Bertie Vidgen and Leon Derczynski | Directions in Abusive Language Training Data: Garbage In, Garbage Out | 26 pages, 5 figures | PLoS ONE 15(12): e0243300 | 10.1371/journal.pone.0243300 | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data-driven analysis and detection of abusive online content covers many
different tasks, phenomena, contexts, and methodologies. This paper
systematically reviews abusive language dataset creation and content in
conjunction with an open website for cataloguing abusive language data. This
collection of knowledge leads to a synthesis providing evidence-based
recommendations for practitioners working with this complex and highly diverse
data.
| [
{
"version": "v1",
"created": "Fri, 3 Apr 2020 16:51:33 GMT"
},
{
"version": "v2",
"created": "Mon, 6 Apr 2020 13:13:12 GMT"
},
{
"version": "v3",
"created": "Mon, 19 Jul 2021 07:40:01 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Vidgen",
"Bertie",
""
],
[
"Derczynski",
"Leon",
""
]
] | TITLE: Directions in Abusive Language Training Data: Garbage In, Garbage Out
ABSTRACT: Data-driven analysis and detection of abusive online content covers many
different tasks, phenomena, contexts, and methodologies. This paper
systematically reviews abusive language dataset creation and content in
conjunction with an open website for cataloguing abusive language data. This
collection of knowledge leads to a synthesis providing evidence-based
recommendations for practitioners working with this complex and highly diverse
data.
| no_new_dataset | 0.946051 |
2005.12390 | Constantinos Siettos | Ioannis Gallos, Evangelos Galaris, Constantinos Siettos | Construction of embedded fMRI resting state functional connectivity
networks using manifold learning | null | Cogn Neurodyn 15, 585-608 (2021) | 10.1007/s11571-020-09645-y | null | q-bio.NC cs.LG cs.NA math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We construct embedded functional connectivity networks (FCN) from benchmark
resting-state functional magnetic resonance imaging (rsfMRI) data acquired from
patients with schizophrenia and healthy controls based on linear and nonlinear
manifold learning algorithms, namely, Multidimensional Scaling (MDS), Isometric
Feature Mapping (ISOMAP) and Diffusion Maps. Furthermore, based on key global
graph-theoretical properties of the embedded FCN, we compare their
classification potential using machine learning techniques. We also assess the
performance of two metrics that are widely used for the construction of FCN
from fMRI, namely the Euclidean distance and the lagged cross-correlation
metric. We show that the FCN constructed with Diffusion Maps and the lagged
cross-correlation metric outperform the other combinations.
| [
{
"version": "v1",
"created": "Mon, 25 May 2020 20:39:29 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Gallos",
"Ioannis",
""
],
[
"Galaris",
"Evangelos",
""
],
[
"Siettos",
"Constantinos",
""
]
] | TITLE: Construction of embedded fMRI resting state functional connectivity
networks using manifold learning
ABSTRACT: We construct embedded functional connectivity networks (FCN) from benchmark
resting-state functional magnetic resonance imaging (rsfMRI) data acquired from
patients with schizophrenia and healthy controls based on linear and nonlinear
manifold learning algorithms, namely, Multidimensional Scaling (MDS), Isometric
Feature Mapping (ISOMAP) and Diffusion Maps. Furthermore, based on key global
graph-theoretical properties of the embedded FCN, we compare their
classification potential using machine learning techniques. We also assess the
performance of two metrics that are widely used for the construction of FCN
from fMRI, namely the Euclidean distance and the lagged cross-correlation
metric. We show that the FCN constructed with Diffusion Maps and the lagged
cross-correlation metric outperform the other combinations.
| no_new_dataset | 0.948298 |
2010.10919 | Yafeng Chen | Yafeng Chen, Wu Guo, Jingjing Shi, Jiajun Qi, Tan Liu | Multi-task Metric Learning for Text-independent Speaker Verification | Not a particularly high-quality work, so we request withdrawal | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we introduce metric learning (ML) to enhance the deep embedding
learning for text-independent speaker verification (SV). Specifically, the deep
speaker embedding network is trained with conventional cross entropy loss and
auxiliary pair-based ML loss function. For the auxiliary ML task, training
samples of a mini-batch are first arranged into pairs, then positive and
negative pairs are selected and weighted through their own and relative
similarities, and finally the auxiliary ML loss is calculated by the similarity
of the selected pairs. To evaluate the proposed method, we conduct experiments
on the Speaker in the Wild (SITW) dataset. The results demonstrate the
effectiveness of the proposed method.
| [
{
"version": "v1",
"created": "Wed, 21 Oct 2020 12:03:36 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 02:28:16 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Chen",
"Yafeng",
""
],
[
"Guo",
"Wu",
""
],
[
"Shi",
"Jingjing",
""
],
[
"Qi",
"Jiajun",
""
],
[
"Liu",
"Tan",
""
]
] | TITLE: Multi-task Metric Learning for Text-independent Speaker Verification
ABSTRACT: In this work, we introduce metric learning (ML) to enhance the deep embedding
learning for text-independent speaker verification (SV). Specifically, the deep
speaker embedding network is trained with conventional cross entropy loss and
auxiliary pair-based ML loss function. For the auxiliary ML task, training
samples of a mini-batch are first arranged into pairs, then positive and
negative pairs are selected and weighted through their own and relative
similarities, and finally the auxiliary ML loss is calculated by the similarity
of the selected pairs. To evaluate the proposed method, we conduct experiments
on the Speaker in the Wild (SITW) dataset. The results demonstrate the
effectiveness of the proposed method.
| no_new_dataset | 0.948537 |
2106.06312 | Zhaomin Wu | Zhaomin Wu, Qinbin Li, Bingsheng He | A Coupled Design of Exploiting Record Similarity for Practical Vertical
Federated Learning | null | Advances in Neural Information Processing Systems 35 (2022)
21087-21100 | null | null | cs.LG cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning is a learning paradigm to enable collaborative learning
across different parties without revealing raw data. Notably, vertical
federated learning (VFL), where parties share the same set of samples but only
hold partial features, has a wide range of real-world applications. However,
most existing studies in VFL disregard the "record linkage" process. They
design algorithms either assuming the data from different parties can be
exactly linked or simply linking each record with its most similar neighboring
record. These approaches may fail to capture the key features from other less
similar records. Moreover, such improper linkage cannot be corrected by
training since existing approaches provide no feedback on linkage during
training. In this paper, we design a novel coupled training paradigm, FedSim,
that integrates one-to-many linkage into the training process. Besides enabling
VFL in many real-world applications with fuzzy identifiers, FedSim also
achieves better performance in traditional VFL tasks. Moreover, we
theoretically analyze the additional privacy risk incurred by sharing
similarities. Our experiments on eight datasets with various similarity metrics
show that FedSim outperforms other state-of-the-art baselines. The codes of
FedSim are available at https://github.com/Xtra-Computing/FedSim.
| [
{
"version": "v1",
"created": "Fri, 11 Jun 2021 11:09:53 GMT"
},
{
"version": "v2",
"created": "Wed, 5 Oct 2022 03:41:47 GMT"
},
{
"version": "v3",
"created": "Sat, 18 Feb 2023 18:30:14 GMT"
},
{
"version": "v4",
"created": "Thu, 23 Mar 2023 16:19:38 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Wu",
"Zhaomin",
""
],
[
"Li",
"Qinbin",
""
],
[
"He",
"Bingsheng",
""
]
] | TITLE: A Coupled Design of Exploiting Record Similarity for Practical Vertical
Federated Learning
ABSTRACT: Federated learning is a learning paradigm to enable collaborative learning
across different parties without revealing raw data. Notably, vertical
federated learning (VFL), where parties share the same set of samples but only
hold partial features, has a wide range of real-world applications. However,
most existing studies in VFL disregard the "record linkage" process. They
design algorithms either assuming the data from different parties can be
exactly linked or simply linking each record with its most similar neighboring
record. These approaches may fail to capture the key features from other less
similar records. Moreover, such improper linkage cannot be corrected by
training since existing approaches provide no feedback on linkage during
training. In this paper, we design a novel coupled training paradigm, FedSim,
that integrates one-to-many linkage into the training process. Besides enabling
VFL in many real-world applications with fuzzy identifiers, FedSim also
achieves better performance in traditional VFL tasks. Moreover, we
theoretically analyze the additional privacy risk incurred by sharing
similarities. Our experiments on eight datasets with various similarity metrics
show that FedSim outperforms other state-of-the-art baselines. The codes of
FedSim are available at https://github.com/Xtra-Computing/FedSim.
| no_new_dataset | 0.946001 |
2108.06518 | Hitoshi Iyatomi | Hayato Arai, Yuto Onga, Kumpei Ikuta, Yusuke Chayama, Hitoshi Iyatomi,
Kenichi Oishi | Disease-oriented image embedding with pseudo-scanner standardization for
content-based image retrieval on 3D brain MRI | 13 pages, 7 figures | null | 10.1109/ACCESS.2021.3129105 | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | To build a robust and practical content-based image retrieval (CBIR) system
that is applicable to a clinical brain MRI database, we propose a new framework
-- Disease-oriented image embedding with pseudo-scanner standardization
(DI-PSS) -- that consists of two core techniques, data harmonization and a
dimension reduction algorithm. Our DI-PSS uses skull stripping and
CycleGAN-based image transformations that map to a standard brain followed by
transformation into a brain image taken with a given reference scanner. Then,
our 3D convolutioinal autoencoders (3D-CAE) with deep metric learning acquires
a low-dimensional embedding that better reflects the characteristics of the
disease. The effectiveness of our proposed framework was tested on the
T1-weighted MRIs selected from the Alzheimer's Disease Neuroimaging Initiative
and the Parkinson's Progression Markers Initiative. We confirmed that our PSS
greatly reduced the variability of low-dimensional embeddings caused by
different scanner and datasets. Compared with the baseline condition, our PSS
reduced the variability in the distance from Alzheimer's disease (AD) to
clinically normal (CN) and Parkinson disease (PD) cases by 15.8-22.6% and
18.0-29.9%, respectively. These properties allow DI-PSS to generate lower
dimensional representations that are more amenable to disease classification.
In AD and CN classification experiments based on spectral clustering, PSS
improved the average accuracy and macro-F1 by 6.2% and 10.7%, respectively.
Given the potential of the DI-PSS for harmonizing images scanned by MRI
scanners that were not used to scan the training data, we expect that the
DI-PSS is suitable for application to a large number of legacy MRIs scanned in
heterogeneous environments.
| [
{
"version": "v1",
"created": "Sat, 14 Aug 2021 11:19:30 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Arai",
"Hayato",
""
],
[
"Onga",
"Yuto",
""
],
[
"Ikuta",
"Kumpei",
""
],
[
"Chayama",
"Yusuke",
""
],
[
"Iyatomi",
"Hitoshi",
""
],
[
"Oishi",
"Kenichi",
""
]
] | TITLE: Disease-oriented image embedding with pseudo-scanner standardization for
content-based image retrieval on 3D brain MRI
ABSTRACT: To build a robust and practical content-based image retrieval (CBIR) system
that is applicable to a clinical brain MRI database, we propose a new framework
-- Disease-oriented image embedding with pseudo-scanner standardization
(DI-PSS) -- that consists of two core techniques, data harmonization and a
dimension reduction algorithm. Our DI-PSS uses skull stripping and
CycleGAN-based image transformations that map to a standard brain followed by
transformation into a brain image taken with a given reference scanner. Then,
our 3D convolutioinal autoencoders (3D-CAE) with deep metric learning acquires
a low-dimensional embedding that better reflects the characteristics of the
disease. The effectiveness of our proposed framework was tested on the
T1-weighted MRIs selected from the Alzheimer's Disease Neuroimaging Initiative
and the Parkinson's Progression Markers Initiative. We confirmed that our PSS
greatly reduced the variability of low-dimensional embeddings caused by
different scanner and datasets. Compared with the baseline condition, our PSS
reduced the variability in the distance from Alzheimer's disease (AD) to
clinically normal (CN) and Parkinson disease (PD) cases by 15.8-22.6% and
18.0-29.9%, respectively. These properties allow DI-PSS to generate lower
dimensional representations that are more amenable to disease classification.
In AD and CN classification experiments based on spectral clustering, PSS
improved the average accuracy and macro-F1 by 6.2% and 10.7%, respectively.
Given the potential of the DI-PSS for harmonizing images scanned by MRI
scanners that were not used to scan the training data, we expect that the
DI-PSS is suitable for application to a large number of legacy MRIs scanned in
heterogeneous environments.
| no_new_dataset | 0.953362 |
2109.04587 | Jeremy Cole | Jeremy R. Cole, Nanjiang Jiang, Panupong Pasupat, Luheng He, Peter
Shaw | Graph-Based Decoding for Task Oriented Semantic Parsing | To appear in EMNLP 5 pages 4 figures | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The dominant paradigm for semantic parsing in recent years is to formulate
parsing as a sequence-to-sequence task, generating predictions with
auto-regressive sequence decoders. In this work, we explore an alternative
paradigm. We formulate semantic parsing as a dependency parsing task, applying
graph-based decoding techniques developed for syntactic parsing. We compare
various decoding techniques given the same pre-trained Transformer encoder on
the TOP dataset, including settings where training data is limited or contains
only partially-annotated examples. We find that our graph-based approach is
competitive with sequence decoders on the standard setting, and offers
significant improvements in data efficiency and settings where
partially-annotated data is available.
| [
{
"version": "v1",
"created": "Thu, 9 Sep 2021 23:22:09 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Cole",
"Jeremy R.",
""
],
[
"Jiang",
"Nanjiang",
""
],
[
"Pasupat",
"Panupong",
""
],
[
"He",
"Luheng",
""
],
[
"Shaw",
"Peter",
""
]
] | TITLE: Graph-Based Decoding for Task Oriented Semantic Parsing
ABSTRACT: The dominant paradigm for semantic parsing in recent years is to formulate
parsing as a sequence-to-sequence task, generating predictions with
auto-regressive sequence decoders. In this work, we explore an alternative
paradigm. We formulate semantic parsing as a dependency parsing task, applying
graph-based decoding techniques developed for syntactic parsing. We compare
various decoding techniques given the same pre-trained Transformer encoder on
the TOP dataset, including settings where training data is limited or contains
only partially-annotated examples. We find that our graph-based approach is
competitive with sequence decoders on the standard setting, and offers
significant improvements in data efficiency and settings where
partially-annotated data is available.
| no_new_dataset | 0.95275 |
2111.15077 | Lei Qi | Lei Qi, Jiaqi Liu, Lei Wang, Yinghuan Shi, Xin Geng | Unsupervised Domain Generalization for Person Re-identification: A
Domain-specific Adaptive Framework | Accepted to Pattern Recognition (PR) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Domain generalization (DG) has attracted much attention in person
re-identification (ReID) recently. It aims to make a model trained on multiple
source domains generalize to an unseen target domain. Although achieving
promising progress, existing methods usually need the source domains to be
labeled, which could be a significant burden for practical ReID tasks. In this
paper, we turn to investigate unsupervised domain generalization for ReID, by
assuming that no label is available for any source domains.
To address this challenging setting, we propose a simple and efficient
domain-specific adaptive framework, and realize it with an adaptive
normalization module designed upon the batch and instance normalization
techniques. In doing so, we successfully yield reliable pseudo-labels to
implement training and also enhance the domain generalization capability of the
model as required. In addition, we show that our framework can even be applied
to improve person ReID under the settings of supervised domain generalization
and unsupervised domain adaptation, demonstrating competitive performance with
respect to relevant methods. Extensive experimental study on benchmark datasets
is conducted to validate the proposed framework. A significance of our work
lies in that it shows the potential of unsupervised domain generalization for
person ReID and sets a strong baseline for the further research on this topic.
| [
{
"version": "v1",
"created": "Tue, 30 Nov 2021 02:35:51 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 07:15:24 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Qi",
"Lei",
""
],
[
"Liu",
"Jiaqi",
""
],
[
"Wang",
"Lei",
""
],
[
"Shi",
"Yinghuan",
""
],
[
"Geng",
"Xin",
""
]
] | TITLE: Unsupervised Domain Generalization for Person Re-identification: A
Domain-specific Adaptive Framework
ABSTRACT: Domain generalization (DG) has attracted much attention in person
re-identification (ReID) recently. It aims to make a model trained on multiple
source domains generalize to an unseen target domain. Although achieving
promising progress, existing methods usually need the source domains to be
labeled, which could be a significant burden for practical ReID tasks. In this
paper, we turn to investigate unsupervised domain generalization for ReID, by
assuming that no label is available for any source domains.
To address this challenging setting, we propose a simple and efficient
domain-specific adaptive framework, and realize it with an adaptive
normalization module designed upon the batch and instance normalization
techniques. In doing so, we successfully yield reliable pseudo-labels to
implement training and also enhance the domain generalization capability of the
model as required. In addition, we show that our framework can even be applied
to improve person ReID under the settings of supervised domain generalization
and unsupervised domain adaptation, demonstrating competitive performance with
respect to relevant methods. Extensive experimental study on benchmark datasets
is conducted to validate the proposed framework. A significance of our work
lies in that it shows the potential of unsupervised domain generalization for
person ReID and sets a strong baseline for the further research on this topic.
| no_new_dataset | 0.944228 |
2112.00995 | Liting Lin | Liting Lin, Heng Fan, Zhipeng Zhang, Yong Xu, Haibin Ling | SwinTrack: A Simple and Strong Baseline for Transformer Tracking | 22 pages, 10 figures | Advances in Neural Information Processing Systems, 2022 | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently Transformer has been largely explored in tracking and shown
state-of-the-art (SOTA) performance. However, existing efforts mainly focus on
fusing and enhancing features generated by convolutional neural networks
(CNNs). The potential of Transformer in representation learning remains
under-explored. In this paper, we aim to further unleash the power of
Transformer by proposing a simple yet efficient fully-attentional tracker,
dubbed SwinTrack, within classic Siamese framework. In particular, both
representation learning and feature fusion in SwinTrack leverage the
Transformer architecture, enabling better feature interactions for tracking
than pure CNN or hybrid CNN-Transformer frameworks. Besides, to further enhance
robustness, we present a novel motion token that embeds historical target
trajectory to improve tracking by providing temporal context. Our motion token
is lightweight with negligible computation but brings clear gains. In our
thorough experiments, SwinTrack exceeds existing approaches on multiple
benchmarks. Particularly, on the challenging LaSOT, SwinTrack sets a new record
with 0.713 SUC score. It also achieves SOTA results on other benchmarks. We
expect SwinTrack to serve as a solid baseline for Transformer tracking and
facilitate future research. Our codes and results are released at
https://github.com/LitingLin/SwinTrack.
| [
{
"version": "v1",
"created": "Thu, 2 Dec 2021 05:56:03 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Dec 2021 15:22:38 GMT"
},
{
"version": "v3",
"created": "Thu, 13 Oct 2022 11:31:03 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Lin",
"Liting",
""
],
[
"Fan",
"Heng",
""
],
[
"Zhang",
"Zhipeng",
""
],
[
"Xu",
"Yong",
""
],
[
"Ling",
"Haibin",
""
]
] | TITLE: SwinTrack: A Simple and Strong Baseline for Transformer Tracking
ABSTRACT: Recently Transformer has been largely explored in tracking and shown
state-of-the-art (SOTA) performance. However, existing efforts mainly focus on
fusing and enhancing features generated by convolutional neural networks
(CNNs). The potential of Transformer in representation learning remains
under-explored. In this paper, we aim to further unleash the power of
Transformer by proposing a simple yet efficient fully-attentional tracker,
dubbed SwinTrack, within classic Siamese framework. In particular, both
representation learning and feature fusion in SwinTrack leverage the
Transformer architecture, enabling better feature interactions for tracking
than pure CNN or hybrid CNN-Transformer frameworks. Besides, to further enhance
robustness, we present a novel motion token that embeds historical target
trajectory to improve tracking by providing temporal context. Our motion token
is lightweight with negligible computation but brings clear gains. In our
thorough experiments, SwinTrack exceeds existing approaches on multiple
benchmarks. Particularly, on the challenging LaSOT, SwinTrack sets a new record
with 0.713 SUC score. It also achieves SOTA results on other benchmarks. We
expect SwinTrack to serve as a solid baseline for Transformer tracking and
facilitate future research. Our codes and results are released at
https://github.com/LitingLin/SwinTrack.
| no_new_dataset | 0.943086 |
2112.11947 | Aizaz Sharif | Aizaz Sharif, Dusica Marijan | Evaluating the Robustness of Deep Reinforcement Learning for Autonomous
Policies in a Multi-agent Urban Driving Environment | null | null | null | null | cs.AI cs.LG cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep reinforcement learning is actively used for training autonomous car
policies in a simulated driving environment. Due to the large availability of
various reinforcement learning algorithms and the lack of their systematic
comparison across different driving scenarios, we are unsure of which ones are
more effective for training autonomous car software in single-agent as well as
multi-agent driving environments. A benchmarking framework for the comparison
of deep reinforcement learning in a vision-based autonomous driving will open
up the possibilities for training better autonomous car driving policies. To
address these challenges, we provide an open and reusable benchmarking
framework for systematic evaluation and comparative analysis of deep
reinforcement learning algorithms for autonomous driving in a single- and
multi-agent environment. Using the framework, we perform a comparative study of
discrete and continuous action space deep reinforcement learning algorithms. We
also propose a comprehensive multi-objective reward function designed for the
evaluation of deep reinforcement learning-based autonomous driving agents. We
run the experiments in a vision-only high-fidelity urban driving simulated
environments. The results indicate that only some of the deep reinforcement
learning algorithms perform consistently better across single and multi-agent
scenarios when trained in various multi-agent-only environment settings. For
example, A3C- and TD3-based autonomous cars perform comparatively better in
terms of more robust actions and minimal driving errors in both single and
multi-agent scenarios. We conclude that different deep reinforcement learning
algorithms exhibit different driving and testing performance in different
scenarios, which underlines the need for their systematic comparative analysis.
The benchmarking framework proposed in this paper facilitates such a
comparison.
| [
{
"version": "v1",
"created": "Wed, 22 Dec 2021 15:14:50 GMT"
},
{
"version": "v2",
"created": "Fri, 27 May 2022 10:53:22 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 17:03:23 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Sharif",
"Aizaz",
""
],
[
"Marijan",
"Dusica",
""
]
] | TITLE: Evaluating the Robustness of Deep Reinforcement Learning for Autonomous
Policies in a Multi-agent Urban Driving Environment
ABSTRACT: Deep reinforcement learning is actively used for training autonomous car
policies in a simulated driving environment. Due to the large availability of
various reinforcement learning algorithms and the lack of their systematic
comparison across different driving scenarios, we are unsure of which ones are
more effective for training autonomous car software in single-agent as well as
multi-agent driving environments. A benchmarking framework for the comparison
of deep reinforcement learning in a vision-based autonomous driving will open
up the possibilities for training better autonomous car driving policies. To
address these challenges, we provide an open and reusable benchmarking
framework for systematic evaluation and comparative analysis of deep
reinforcement learning algorithms for autonomous driving in a single- and
multi-agent environment. Using the framework, we perform a comparative study of
discrete and continuous action space deep reinforcement learning algorithms. We
also propose a comprehensive multi-objective reward function designed for the
evaluation of deep reinforcement learning-based autonomous driving agents. We
run the experiments in a vision-only high-fidelity urban driving simulated
environments. The results indicate that only some of the deep reinforcement
learning algorithms perform consistently better across single and multi-agent
scenarios when trained in various multi-agent-only environment settings. For
example, A3C- and TD3-based autonomous cars perform comparatively better in
terms of more robust actions and minimal driving errors in both single and
multi-agent scenarios. We conclude that different deep reinforcement learning
algorithms exhibit different driving and testing performance in different
scenarios, which underlines the need for their systematic comparative analysis.
The benchmarking framework proposed in this paper facilitates such a
comparison.
| no_new_dataset | 0.947624 |
2202.07291 | Sangjin Lee | Sangjin Lee, Hyeongmin Lee, Chajin Shin, Hanbin Son, Sangyoun Lee | Exploring Discontinuity for Video Frame Interpolation | highlight at CVPR 2023 (10% of accepted papers) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video frame interpolation (VFI) is the task that synthesizes the intermediate
frame given two consecutive frames. Most of the previous studies have focused
on appropriate frame warping operations and refinement modules for the warped
frames. These studies have been conducted on natural videos containing only
continuous motions. However, many practical videos contain various unnatural
objects with discontinuous motions such as logos, user interfaces and
subtitles. We propose three techniques to make the existing deep learning-based
VFI architectures robust to these elements. First is a novel data augmentation
strategy called figure-text mixing (FTM) which can make the models learn
discontinuous motions during training stage without any extra dataset. Second,
we propose a simple but effective module that predicts a map called
discontinuity map (D-map), which densely distinguishes between areas of
continuous and discontinuous motions. Lastly, we propose loss functions to give
supervisions of the discontinuous motion areas which can be applied along with
FTM and D-map. We additionally collect a special test benchmark called
Graphical Discontinuous Motion (GDM) dataset consisting of some mobile games
and chatting videos. Applied to the various state-of-the-art VFI networks, our
method significantly improves the interpolation qualities on the videos from
not only GDM dataset, but also the existing benchmarks containing only
continuous motions such as Vimeo90K, UCF101, and DAVIS.
| [
{
"version": "v1",
"created": "Tue, 15 Feb 2022 10:17:02 GMT"
},
{
"version": "v2",
"created": "Thu, 17 Feb 2022 02:49:52 GMT"
},
{
"version": "v3",
"created": "Thu, 8 Dec 2022 07:02:29 GMT"
},
{
"version": "v4",
"created": "Fri, 10 Mar 2023 06:26:10 GMT"
},
{
"version": "v5",
"created": "Thu, 23 Mar 2023 04:44:42 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Lee",
"Sangjin",
""
],
[
"Lee",
"Hyeongmin",
""
],
[
"Shin",
"Chajin",
""
],
[
"Son",
"Hanbin",
""
],
[
"Lee",
"Sangyoun",
""
]
] | TITLE: Exploring Discontinuity for Video Frame Interpolation
ABSTRACT: Video frame interpolation (VFI) is the task that synthesizes the intermediate
frame given two consecutive frames. Most of the previous studies have focused
on appropriate frame warping operations and refinement modules for the warped
frames. These studies have been conducted on natural videos containing only
continuous motions. However, many practical videos contain various unnatural
objects with discontinuous motions such as logos, user interfaces and
subtitles. We propose three techniques to make the existing deep learning-based
VFI architectures robust to these elements. First is a novel data augmentation
strategy called figure-text mixing (FTM) which can make the models learn
discontinuous motions during training stage without any extra dataset. Second,
we propose a simple but effective module that predicts a map called
discontinuity map (D-map), which densely distinguishes between areas of
continuous and discontinuous motions. Lastly, we propose loss functions to give
supervisions of the discontinuous motion areas which can be applied along with
FTM and D-map. We additionally collect a special test benchmark called
Graphical Discontinuous Motion (GDM) dataset consisting of some mobile games
and chatting videos. Applied to the various state-of-the-art VFI networks, our
method significantly improves the interpolation qualities on the videos from
not only GDM dataset, but also the existing benchmarks containing only
continuous motions such as Vimeo90K, UCF101, and DAVIS.
| new_dataset | 0.970576 |
2203.04287 | Fangyun Wei | Yutong Chen, Fangyun Wei, Xiao Sun, Zhirong Wu, Stephen Lin | A Simple Multi-Modality Transfer Learning Baseline for Sign Language
Translation | Accepted by CVPR 2022. 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 paper proposes a simple transfer learning baseline for sign language
translation. Existing sign language datasets (e.g. PHOENIX-2014T, CSL-Daily)
contain only about 10K-20K pairs of sign videos, gloss annotations and texts,
which are an order of magnitude smaller than typical parallel data for training
spoken language translation models. Data is thus a bottleneck for training
effective sign language translation models. To mitigate this problem, we
propose to progressively pretrain the model from general-domain datasets that
include a large amount of external supervision to within-domain datasets.
Concretely, we pretrain the sign-to-gloss visual network on the general domain
of human actions and the within-domain of a sign-to-gloss dataset, and pretrain
the gloss-to-text translation network on the general domain of a multilingual
corpus and the within-domain of a gloss-to-text corpus. The joint model is
fine-tuned with an additional module named the visual-language mapper that
connects the two networks. This simple baseline surpasses the previous
state-of-the-art results on two sign language translation benchmarks,
demonstrating the effectiveness of transfer learning. With its simplicity and
strong performance, this approach can serve as a solid baseline for future
research. Code and models are available at: https://github.com/FangyunWei/SLRT.
| [
{
"version": "v1",
"created": "Tue, 8 Mar 2022 18:59:56 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 02:54:23 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Chen",
"Yutong",
""
],
[
"Wei",
"Fangyun",
""
],
[
"Sun",
"Xiao",
""
],
[
"Wu",
"Zhirong",
""
],
[
"Lin",
"Stephen",
""
]
] | TITLE: A Simple Multi-Modality Transfer Learning Baseline for Sign Language
Translation
ABSTRACT: This paper proposes a simple transfer learning baseline for sign language
translation. Existing sign language datasets (e.g. PHOENIX-2014T, CSL-Daily)
contain only about 10K-20K pairs of sign videos, gloss annotations and texts,
which are an order of magnitude smaller than typical parallel data for training
spoken language translation models. Data is thus a bottleneck for training
effective sign language translation models. To mitigate this problem, we
propose to progressively pretrain the model from general-domain datasets that
include a large amount of external supervision to within-domain datasets.
Concretely, we pretrain the sign-to-gloss visual network on the general domain
of human actions and the within-domain of a sign-to-gloss dataset, and pretrain
the gloss-to-text translation network on the general domain of a multilingual
corpus and the within-domain of a gloss-to-text corpus. The joint model is
fine-tuned with an additional module named the visual-language mapper that
connects the two networks. This simple baseline surpasses the previous
state-of-the-art results on two sign language translation benchmarks,
demonstrating the effectiveness of transfer learning. With its simplicity and
strong performance, this approach can serve as a solid baseline for future
research. Code and models are available at: https://github.com/FangyunWei/SLRT.
| no_new_dataset | 0.944125 |
2204.03722 | Gustavo Olague Dr. | Gustavo Olague, Jose Armando Menendez-Clavijo, Matthieu Olague, Arturo
Ocampo, Gerardo Ibarra-Vazquez, Rocio Ochoa and Roberto Pineda | Automated Design of Salient Object Detection Algorithms with Brain
Programming | 35 pages, 5 figures | https://www.mdpi.com/journal/applsci 2020 | 10.3390/app122010686 | https://doi.org/10.3390/app122010686 | cs.CV cs.LG cs.NE cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Despite recent improvements in computer vision, artificial visual systems'
design is still daunting since an explanation of visual computing algorithms
remains elusive. Salient object detection is one problem that is still open due
to the difficulty of understanding the brain's inner workings. Progress on this
research area follows the traditional path of hand-made designs using
neuroscience knowledge. In recent years two different approaches based on
genetic programming appear to enhance their technique. One follows the idea of
combining previous hand-made methods through genetic programming and fuzzy
logic. The other approach consists of improving the inner computational
structures of basic hand-made models through artificial evolution. This
research work proposes expanding the artificial dorsal stream using a recent
proposal to solve salient object detection problems. This approach uses the
benefits of the two main aspects of this research area: fixation prediction and
detection of salient objects. We decided to apply the fusion of visual saliency
and image segmentation algorithms as a template. The proposed methodology
discovers several critical structures in the template through artificial
evolution. We present results on a benchmark designed by experts with
outstanding results in comparison with the state-of-the-art.
| [
{
"version": "v1",
"created": "Thu, 7 Apr 2022 20:21:30 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Olague",
"Gustavo",
""
],
[
"Menendez-Clavijo",
"Jose Armando",
""
],
[
"Olague",
"Matthieu",
""
],
[
"Ocampo",
"Arturo",
""
],
[
"Ibarra-Vazquez",
"Gerardo",
""
],
[
"Ochoa",
"Rocio",
""
],
[
"Pineda",
"Roberto",
""
]
] | TITLE: Automated Design of Salient Object Detection Algorithms with Brain
Programming
ABSTRACT: Despite recent improvements in computer vision, artificial visual systems'
design is still daunting since an explanation of visual computing algorithms
remains elusive. Salient object detection is one problem that is still open due
to the difficulty of understanding the brain's inner workings. Progress on this
research area follows the traditional path of hand-made designs using
neuroscience knowledge. In recent years two different approaches based on
genetic programming appear to enhance their technique. One follows the idea of
combining previous hand-made methods through genetic programming and fuzzy
logic. The other approach consists of improving the inner computational
structures of basic hand-made models through artificial evolution. This
research work proposes expanding the artificial dorsal stream using a recent
proposal to solve salient object detection problems. This approach uses the
benefits of the two main aspects of this research area: fixation prediction and
detection of salient objects. We decided to apply the fusion of visual saliency
and image segmentation algorithms as a template. The proposed methodology
discovers several critical structures in the template through artificial
evolution. We present results on a benchmark designed by experts with
outstanding results in comparison with the state-of-the-art.
| no_new_dataset | 0.943712 |
2204.08696 | Guangwei Gao | Guangwei Gao, Zixiang Xu, Juncheng Li, Jian Yang, Tieyong Zeng and
Guo-Jun Qi | CTCNet: A CNN-Transformer Cooperation Network for Face Image
Super-Resolution | IEEE Transactions on Image Processing, 12 figures, 9 tables | null | 10.1109/TIP.2023.3261747 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, deep convolution neural networks (CNNs) steered face
super-resolution methods have achieved great progress in restoring degraded
facial details by jointly training with facial priors. However, these methods
have some obvious limitations. On the one hand, multi-task joint learning
requires additional marking on the dataset, and the introduced prior network
will significantly increase the computational cost of the model. On the other
hand, the limited receptive field of CNN will reduce the fidelity and
naturalness of the reconstructed facial images, resulting in suboptimal
reconstructed images. In this work, we propose an efficient CNN-Transformer
Cooperation Network (CTCNet) for face super-resolution tasks, which uses the
multi-scale connected encoder-decoder architecture as the backbone.
Specifically, we first devise a novel Local-Global Feature Cooperation Module
(LGCM), which is composed of a Facial Structure Attention Unit (FSAU) and a
Transformer block, to promote the consistency of local facial detail and global
facial structure restoration simultaneously. Then, we design an efficient
Feature Refinement Module (FRM) to enhance the encoded features. Finally, to
further improve the restoration of fine facial details, we present a
Multi-scale Feature Fusion Unit (MFFU) to adaptively fuse the features from
different stages in the encoder procedure. Extensive evaluations on various
datasets have assessed that the proposed CTCNet can outperform other
state-of-the-art methods significantly. Source code will be available at
https://github.com/IVIPLab/CTCNet.
| [
{
"version": "v1",
"created": "Tue, 19 Apr 2022 06:38:29 GMT"
},
{
"version": "v2",
"created": "Mon, 30 Jan 2023 01:51:04 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 09:44:22 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Gao",
"Guangwei",
""
],
[
"Xu",
"Zixiang",
""
],
[
"Li",
"Juncheng",
""
],
[
"Yang",
"Jian",
""
],
[
"Zeng",
"Tieyong",
""
],
[
"Qi",
"Guo-Jun",
""
]
] | TITLE: CTCNet: A CNN-Transformer Cooperation Network for Face Image
Super-Resolution
ABSTRACT: Recently, deep convolution neural networks (CNNs) steered face
super-resolution methods have achieved great progress in restoring degraded
facial details by jointly training with facial priors. However, these methods
have some obvious limitations. On the one hand, multi-task joint learning
requires additional marking on the dataset, and the introduced prior network
will significantly increase the computational cost of the model. On the other
hand, the limited receptive field of CNN will reduce the fidelity and
naturalness of the reconstructed facial images, resulting in suboptimal
reconstructed images. In this work, we propose an efficient CNN-Transformer
Cooperation Network (CTCNet) for face super-resolution tasks, which uses the
multi-scale connected encoder-decoder architecture as the backbone.
Specifically, we first devise a novel Local-Global Feature Cooperation Module
(LGCM), which is composed of a Facial Structure Attention Unit (FSAU) and a
Transformer block, to promote the consistency of local facial detail and global
facial structure restoration simultaneously. Then, we design an efficient
Feature Refinement Module (FRM) to enhance the encoded features. Finally, to
further improve the restoration of fine facial details, we present a
Multi-scale Feature Fusion Unit (MFFU) to adaptively fuse the features from
different stages in the encoder procedure. Extensive evaluations on various
datasets have assessed that the proposed CTCNet can outperform other
state-of-the-art methods significantly. Source code will be available at
https://github.com/IVIPLab/CTCNet.
| no_new_dataset | 0.946399 |
2205.07680 | Bin Liu | Bo Li, Kaitao Xue, Bin Liu, Yu-Kun Lai | BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models | 18 pages, 13 figures | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image-to-image translation is an important and challenging problem in
computer vision and image processing. Diffusion models (DM) have shown great
potentials for high-quality image synthesis, and have gained competitive
performance on the task of image-to-image translation. However, most of the
existing diffusion models treat image-to-image translation as conditional
generation processes, and suffer heavily from the gap between distinct domains.
In this paper, a novel image-to-image translation method based on the Brownian
Bridge Diffusion Model (BBDM) is proposed, which models image-to-image
translation as a stochastic Brownian bridge process, and learns the translation
between two domains directly through the bidirectional diffusion process rather
than a conditional generation process. To the best of our knowledge, it is the
first work that proposes Brownian Bridge diffusion process for image-to-image
translation. Experimental results on various benchmarks demonstrate that the
proposed BBDM model achieves competitive performance through both visual
inspection and measurable metrics.
| [
{
"version": "v1",
"created": "Mon, 16 May 2022 13:47:02 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 08:29:12 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Li",
"Bo",
""
],
[
"Xue",
"Kaitao",
""
],
[
"Liu",
"Bin",
""
],
[
"Lai",
"Yu-Kun",
""
]
] | TITLE: BBDM: Image-to-image Translation with Brownian Bridge Diffusion Models
ABSTRACT: Image-to-image translation is an important and challenging problem in
computer vision and image processing. Diffusion models (DM) have shown great
potentials for high-quality image synthesis, and have gained competitive
performance on the task of image-to-image translation. However, most of the
existing diffusion models treat image-to-image translation as conditional
generation processes, and suffer heavily from the gap between distinct domains.
In this paper, a novel image-to-image translation method based on the Brownian
Bridge Diffusion Model (BBDM) is proposed, which models image-to-image
translation as a stochastic Brownian bridge process, and learns the translation
between two domains directly through the bidirectional diffusion process rather
than a conditional generation process. To the best of our knowledge, it is the
first work that proposes Brownian Bridge diffusion process for image-to-image
translation. Experimental results on various benchmarks demonstrate that the
proposed BBDM model achieves competitive performance through both visual
inspection and measurable metrics.
| no_new_dataset | 0.955194 |
2205.15448 | Ce Zheng | Ce Zheng, Matias Mendieta, Taojiannan Yang, Guo-Jun Qi, Chen Chen | FeatER: An Efficient Network for Human Reconstruction via Feature
Map-Based TransformER | CVPR 2023 | null | null | null | cs.CV cs.AI cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, vision transformers have shown great success in a set of human
reconstruction tasks such as 2D human pose estimation (2D HPE), 3D human pose
estimation (3D HPE), and human mesh reconstruction (HMR) tasks. In these tasks,
feature map representations of the human structural information are often
extracted first from the image by a CNN (such as HRNet), and then further
processed by transformer to predict the heatmaps (encodes each joint's location
into a feature map with a Gaussian distribution) for HPE or HMR. However,
existing transformer architectures are not able to process these feature map
inputs directly, forcing an unnatural flattening of the location-sensitive
human structural information. Furthermore, much of the performance benefit in
recent HPE and HMR methods has come at the cost of ever-increasing computation
and memory needs. Therefore, to simultaneously address these problems, we
propose FeatER, a novel transformer design that preserves the inherent
structure of feature map representations when modeling attention while reducing
memory and computational costs. Taking advantage of FeatER, we build an
efficient network for a set of human reconstruction tasks including 2D HPE, 3D
HPE, and HMR. A feature map reconstruction module is applied to improve the
performance of the estimated human pose and mesh. Extensive experiments
demonstrate the effectiveness of FeatER on various human pose and mesh
datasets. For instance, FeatER outperforms the SOTA method MeshGraphormer by
requiring 5% of Params and 16% of MACs on Human3.6M and 3DPW datasets. The
project webpage is https://zczcwh.github.io/feater_page/.
| [
{
"version": "v1",
"created": "Mon, 30 May 2022 22:09:57 GMT"
},
{
"version": "v2",
"created": "Wed, 23 Nov 2022 00:03:20 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 15:48:05 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Zheng",
"Ce",
""
],
[
"Mendieta",
"Matias",
""
],
[
"Yang",
"Taojiannan",
""
],
[
"Qi",
"Guo-Jun",
""
],
[
"Chen",
"Chen",
""
]
] | TITLE: FeatER: An Efficient Network for Human Reconstruction via Feature
Map-Based TransformER
ABSTRACT: Recently, vision transformers have shown great success in a set of human
reconstruction tasks such as 2D human pose estimation (2D HPE), 3D human pose
estimation (3D HPE), and human mesh reconstruction (HMR) tasks. In these tasks,
feature map representations of the human structural information are often
extracted first from the image by a CNN (such as HRNet), and then further
processed by transformer to predict the heatmaps (encodes each joint's location
into a feature map with a Gaussian distribution) for HPE or HMR. However,
existing transformer architectures are not able to process these feature map
inputs directly, forcing an unnatural flattening of the location-sensitive
human structural information. Furthermore, much of the performance benefit in
recent HPE and HMR methods has come at the cost of ever-increasing computation
and memory needs. Therefore, to simultaneously address these problems, we
propose FeatER, a novel transformer design that preserves the inherent
structure of feature map representations when modeling attention while reducing
memory and computational costs. Taking advantage of FeatER, we build an
efficient network for a set of human reconstruction tasks including 2D HPE, 3D
HPE, and HMR. A feature map reconstruction module is applied to improve the
performance of the estimated human pose and mesh. Extensive experiments
demonstrate the effectiveness of FeatER on various human pose and mesh
datasets. For instance, FeatER outperforms the SOTA method MeshGraphormer by
requiring 5% of Params and 16% of MACs on Human3.6M and 3DPW datasets. The
project webpage is https://zczcwh.github.io/feater_page/.
| no_new_dataset | 0.949012 |
2206.06067 | Xinzhu Ma | Zengyu Qiu, Xinzhu Ma, Kunlin Yang, Chunya Liu, Jun Hou, Shuai Yi,
Wanli Ouyang | Better Teacher Better Student: Dynamic Prior Knowledge for Knowledge
Distillation | ICLR'23 accepted | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Knowledge distillation (KD) has shown very promising capabilities in
transferring learning representations from large models (teachers) to small
models (students). However, as the capacity gap between students and teachers
becomes larger, existing KD methods fail to achieve better results. Our work
shows that the `prior knowledge' is vital to KD, especially when applying large
teachers. Particularly, we propose the dynamic prior knowledge (DPK), which
integrates part of teacher's features as the prior knowledge before the feature
distillation. This means that our method also takes the teacher's feature as
`input', not just `target'. Besides, we dynamically adjust the ratio of the
prior knowledge during the training phase according to the feature gap, thus
guiding the student in an appropriate difficulty. To evaluate the proposed
method, we conduct extensive experiments on two image classification benchmarks
(i.e. CIFAR100 and ImageNet) and an object detection benchmark (i.e. MS COCO.
The results demonstrate the superiority of our method in performance under
varying settings. Besides, our DPK makes the performance of the student model
positively correlated with that of the teacher model, which means that we can
further boost the accuracy of students by applying larger teachers. More
importantly, DPK provides a fast solution in teacher model selection for any
given model.
| [
{
"version": "v1",
"created": "Mon, 13 Jun 2022 11:52:13 GMT"
},
{
"version": "v2",
"created": "Tue, 14 Jun 2022 05:59:40 GMT"
},
{
"version": "v3",
"created": "Thu, 16 Feb 2023 12:37:43 GMT"
},
{
"version": "v4",
"created": "Thu, 23 Mar 2023 11:17:31 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Qiu",
"Zengyu",
""
],
[
"Ma",
"Xinzhu",
""
],
[
"Yang",
"Kunlin",
""
],
[
"Liu",
"Chunya",
""
],
[
"Hou",
"Jun",
""
],
[
"Yi",
"Shuai",
""
],
[
"Ouyang",
"Wanli",
""
]
] | TITLE: Better Teacher Better Student: Dynamic Prior Knowledge for Knowledge
Distillation
ABSTRACT: Knowledge distillation (KD) has shown very promising capabilities in
transferring learning representations from large models (teachers) to small
models (students). However, as the capacity gap between students and teachers
becomes larger, existing KD methods fail to achieve better results. Our work
shows that the `prior knowledge' is vital to KD, especially when applying large
teachers. Particularly, we propose the dynamic prior knowledge (DPK), which
integrates part of teacher's features as the prior knowledge before the feature
distillation. This means that our method also takes the teacher's feature as
`input', not just `target'. Besides, we dynamically adjust the ratio of the
prior knowledge during the training phase according to the feature gap, thus
guiding the student in an appropriate difficulty. To evaluate the proposed
method, we conduct extensive experiments on two image classification benchmarks
(i.e. CIFAR100 and ImageNet) and an object detection benchmark (i.e. MS COCO.
The results demonstrate the superiority of our method in performance under
varying settings. Besides, our DPK makes the performance of the student model
positively correlated with that of the teacher model, which means that we can
further boost the accuracy of students by applying larger teachers. More
importantly, DPK provides a fast solution in teacher model selection for any
given model.
| no_new_dataset | 0.945551 |
2206.08289 | Shengsen Wu | Shengsen Wu, Yan Bai, Yihang Lou, Xiongkun Linghu, Jianzhong He and
Ling-Yu Duan | Switchable Representation Learning Framework with Self-compatibility | Accepted by CVPR 2023 | null | null | null | cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Real-world visual search systems involve deployments on multiple platforms
with different computing and storage resources. Deploying a unified model that
suits the minimal-constrain platforms leads to limited accuracy. It is expected
to deploy models with different capacities adapting to the resource
constraints, which requires features extracted by these models to be aligned in
the metric space. The method to achieve feature alignments is called
``compatible learning''. Existing research mainly focuses on the one-to-one
compatible paradigm, which is limited in learning compatibility among multiple
models. We propose a Switchable representation learning Framework with
Self-Compatibility (SFSC). SFSC generates a series of compatible sub-models
with different capacities through one training process. The optimization of
sub-models faces gradients conflict, and we mitigate this problem from the
perspective of the magnitude and direction. We adjust the priorities of
sub-models dynamically through uncertainty estimation to co-optimize sub-models
properly. Besides, the gradients with conflicting directions are projected to
avoid mutual interference. SFSC achieves state-of-the-art performance on the
evaluated datasets.
| [
{
"version": "v1",
"created": "Thu, 16 Jun 2022 16:46:32 GMT"
},
{
"version": "v2",
"created": "Mon, 6 Mar 2023 03:54:50 GMT"
},
{
"version": "v3",
"created": "Wed, 8 Mar 2023 06:00:29 GMT"
},
{
"version": "v4",
"created": "Thu, 23 Mar 2023 10:54:32 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Wu",
"Shengsen",
""
],
[
"Bai",
"Yan",
""
],
[
"Lou",
"Yihang",
""
],
[
"Linghu",
"Xiongkun",
""
],
[
"He",
"Jianzhong",
""
],
[
"Duan",
"Ling-Yu",
""
]
] | TITLE: Switchable Representation Learning Framework with Self-compatibility
ABSTRACT: Real-world visual search systems involve deployments on multiple platforms
with different computing and storage resources. Deploying a unified model that
suits the minimal-constrain platforms leads to limited accuracy. It is expected
to deploy models with different capacities adapting to the resource
constraints, which requires features extracted by these models to be aligned in
the metric space. The method to achieve feature alignments is called
``compatible learning''. Existing research mainly focuses on the one-to-one
compatible paradigm, which is limited in learning compatibility among multiple
models. We propose a Switchable representation learning Framework with
Self-Compatibility (SFSC). SFSC generates a series of compatible sub-models
with different capacities through one training process. The optimization of
sub-models faces gradients conflict, and we mitigate this problem from the
perspective of the magnitude and direction. We adjust the priorities of
sub-models dynamically through uncertainty estimation to co-optimize sub-models
properly. Besides, the gradients with conflicting directions are projected to
avoid mutual interference. SFSC achieves state-of-the-art performance on the
evaluated datasets.
| no_new_dataset | 0.946646 |
2206.11488 | Hong-You Chen | Hong-You Chen, Cheng-Hao Tu, Ziwei Li, Han-Wei Shen, Wei-Lun Chao | On the Importance and Applicability of Pre-Training for Federated
Learning | Accepted to ICLR 2023 | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Pre-training is prevalent in nowadays deep learning to improve the learned
model's performance. However, in the literature on federated learning (FL),
neural networks are mostly initialized with random weights. These attract our
interest in conducting a systematic study to explore pre-training for FL.
Across multiple visual recognition benchmarks, we found that pre-training can
not only improve FL, but also close its accuracy gap to the counterpart
centralized learning, especially in the challenging cases of non-IID clients'
data. To make our findings applicable to situations where pre-trained models
are not directly available, we explore pre-training with synthetic data or even
with clients' data in a decentralized manner, and found that they can already
improve FL notably. Interestingly, many of the techniques we explore are
complementary to each other to further boost the performance, and we view this
as a critical result toward scaling up deep FL for real-world applications. We
conclude our paper with an attempt to understand the effect of pre-training on
FL. We found that pre-training enables the learned global models under
different clients' data conditions to converge to the same loss basin, and
makes global aggregation in FL more stable. Nevertheless, pre-training seems to
not alleviate local model drifting, a fundamental problem in FL under non-IID
data.
| [
{
"version": "v1",
"created": "Thu, 23 Jun 2022 06:02:33 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Oct 2022 17:30:15 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 03:27:40 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Chen",
"Hong-You",
""
],
[
"Tu",
"Cheng-Hao",
""
],
[
"Li",
"Ziwei",
""
],
[
"Shen",
"Han-Wei",
""
],
[
"Chao",
"Wei-Lun",
""
]
] | TITLE: On the Importance and Applicability of Pre-Training for Federated
Learning
ABSTRACT: Pre-training is prevalent in nowadays deep learning to improve the learned
model's performance. However, in the literature on federated learning (FL),
neural networks are mostly initialized with random weights. These attract our
interest in conducting a systematic study to explore pre-training for FL.
Across multiple visual recognition benchmarks, we found that pre-training can
not only improve FL, but also close its accuracy gap to the counterpart
centralized learning, especially in the challenging cases of non-IID clients'
data. To make our findings applicable to situations where pre-trained models
are not directly available, we explore pre-training with synthetic data or even
with clients' data in a decentralized manner, and found that they can already
improve FL notably. Interestingly, many of the techniques we explore are
complementary to each other to further boost the performance, and we view this
as a critical result toward scaling up deep FL for real-world applications. We
conclude our paper with an attempt to understand the effect of pre-training on
FL. We found that pre-training enables the learned global models under
different clients' data conditions to converge to the same loss basin, and
makes global aggregation in FL more stable. Nevertheless, pre-training seems to
not alleviate local model drifting, a fundamental problem in FL under non-IID
data.
| no_new_dataset | 0.944382 |
2206.14615 | Xu Wu | Mahmoud Yaseen, Xu Wu | Quantification of Deep Neural Network Prediction Uncertainties for VVUQ
of Machine Learning Models | 33 pages, 4 tables, 13 figures | null | 10.1080/00295639.2022.2123203 | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent performance breakthroughs in Artificial intelligence (AI) and Machine
learning (ML), especially advances in Deep learning (DL), the availability of
powerful, easy-to-use ML libraries (e.g., scikit-learn, TensorFlow, PyTorch.),
and increasing computational power have led to unprecedented interest in AI/ML
among nuclear engineers. For physics-based computational models, Verification,
Validation and Uncertainty Quantification (VVUQ) have been very widely
investigated and a lot of methodologies have been developed. However, VVUQ of
ML models has been relatively less studied, especially in nuclear engineering.
In this work, we focus on UQ of ML models as a preliminary step of ML VVUQ,
more specifically, Deep Neural Networks (DNNs) because they are the most widely
used supervised ML algorithm for both regression and classification tasks. This
work aims at quantifying the prediction, or approximation uncertainties of DNNs
when they are used as surrogate models for expensive physical models. Three
techniques for UQ of DNNs are compared, namely Monte Carlo Dropout (MCD), Deep
Ensembles (DE) and Bayesian Neural Networks (BNNs). Two nuclear engineering
examples are used to benchmark these methods, (1) time-dependent fission gas
release data using the Bison code, and (2) void fraction simulation based on
the BFBT benchmark using the TRACE code. It was found that the three methods
typically require different DNN architectures and hyperparameters to optimize
their performance. The UQ results also depend on the amount of training data
available and the nature of the data. Overall, all these three methods can
provide reasonable estimations of the approximation uncertainties. The
uncertainties are generally smaller when the mean predictions are close to the
test data, while the BNN methods usually produce larger uncertainties than MCD
and DE.
| [
{
"version": "v1",
"created": "Mon, 27 Jun 2022 20:49:57 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Yaseen",
"Mahmoud",
""
],
[
"Wu",
"Xu",
""
]
] | TITLE: Quantification of Deep Neural Network Prediction Uncertainties for VVUQ
of Machine Learning Models
ABSTRACT: Recent performance breakthroughs in Artificial intelligence (AI) and Machine
learning (ML), especially advances in Deep learning (DL), the availability of
powerful, easy-to-use ML libraries (e.g., scikit-learn, TensorFlow, PyTorch.),
and increasing computational power have led to unprecedented interest in AI/ML
among nuclear engineers. For physics-based computational models, Verification,
Validation and Uncertainty Quantification (VVUQ) have been very widely
investigated and a lot of methodologies have been developed. However, VVUQ of
ML models has been relatively less studied, especially in nuclear engineering.
In this work, we focus on UQ of ML models as a preliminary step of ML VVUQ,
more specifically, Deep Neural Networks (DNNs) because they are the most widely
used supervised ML algorithm for both regression and classification tasks. This
work aims at quantifying the prediction, or approximation uncertainties of DNNs
when they are used as surrogate models for expensive physical models. Three
techniques for UQ of DNNs are compared, namely Monte Carlo Dropout (MCD), Deep
Ensembles (DE) and Bayesian Neural Networks (BNNs). Two nuclear engineering
examples are used to benchmark these methods, (1) time-dependent fission gas
release data using the Bison code, and (2) void fraction simulation based on
the BFBT benchmark using the TRACE code. It was found that the three methods
typically require different DNN architectures and hyperparameters to optimize
their performance. The UQ results also depend on the amount of training data
available and the nature of the data. Overall, all these three methods can
provide reasonable estimations of the approximation uncertainties. The
uncertainties are generally smaller when the mean predictions are close to the
test data, while the BNN methods usually produce larger uncertainties than MCD
and DE.
| no_new_dataset | 0.939582 |
2207.11020 | Tomas Kulvicius | Peter B Marschik, Tomas Kulvicius, Sarah Fl\"ugge, Claudius Widmann,
Karin Nielsen-Saines, Martin Schulte-R\"uther, Britta H\"uning, Sven B\"olte,
Luise Poustka, Jeff Sigafoos, Florentin W\"org\"otter, Christa Einspieler,
Dajie Zhang | Open video data sharing in developmental and behavioural science | null | iScience, 2023 | 10.1016/j.isci.2023.106348 | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Video recording is a widely used method for documenting infant and child
behaviours in research and clinical practice. Video data has rarely been shared
due to ethical concerns of confidentiality, although the need of shared
large-scaled datasets remains increasing. This demand is even more imperative
when data-driven computer-based approaches are involved, such as screening
tools to complement clinical assessments. To share data while abiding by
privacy protection rules, a critical question arises whether efforts at data
de-identification reduce data utility? We addressed this question by showcasing
the Prechtl's general movements assessment (GMA), an established and globally
practised video-based diagnostic tool in early infancy for detecting
neurological deficits, such as cerebral palsy. To date, no shared
expert-annotated large data repositories for infant movement analyses exist.
Such datasets would massively benefit training and recalibration of human
assessors and the development of computer-based approaches. In the current
study, sequences from a prospective longitudinal infant cohort with a total of
19451 available general movements video snippets were randomly selected for
human clinical reasoning and computer-based analysis. We demonstrated for the
first time that pseudonymisation by face-blurring video recordings is a viable
approach. The video redaction did not affect classification accuracy for either
human assessors or computer vision methods, suggesting an adequate and
easy-to-apply solution for sharing movement video data. We call for further
explorations into efficient and privacy rule-conforming approaches for
deidentifying video data in scientific and clinical fields beyond movement
assessments. These approaches shall enable sharing and merging stand-alone
video datasets into large data pools to advance science and public health.
| [
{
"version": "v1",
"created": "Fri, 22 Jul 2022 11:47:47 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Marschik",
"Peter B",
""
],
[
"Kulvicius",
"Tomas",
""
],
[
"Flügge",
"Sarah",
""
],
[
"Widmann",
"Claudius",
""
],
[
"Nielsen-Saines",
"Karin",
""
],
[
"Schulte-Rüther",
"Martin",
""
],
[
"Hüning",
"Britta",
""
],
[
"Bölte",
"Sven",
""
],
[
"Poustka",
"Luise",
""
],
[
"Sigafoos",
"Jeff",
""
],
[
"Wörgötter",
"Florentin",
""
],
[
"Einspieler",
"Christa",
""
],
[
"Zhang",
"Dajie",
""
]
] | TITLE: Open video data sharing in developmental and behavioural science
ABSTRACT: Video recording is a widely used method for documenting infant and child
behaviours in research and clinical practice. Video data has rarely been shared
due to ethical concerns of confidentiality, although the need of shared
large-scaled datasets remains increasing. This demand is even more imperative
when data-driven computer-based approaches are involved, such as screening
tools to complement clinical assessments. To share data while abiding by
privacy protection rules, a critical question arises whether efforts at data
de-identification reduce data utility? We addressed this question by showcasing
the Prechtl's general movements assessment (GMA), an established and globally
practised video-based diagnostic tool in early infancy for detecting
neurological deficits, such as cerebral palsy. To date, no shared
expert-annotated large data repositories for infant movement analyses exist.
Such datasets would massively benefit training and recalibration of human
assessors and the development of computer-based approaches. In the current
study, sequences from a prospective longitudinal infant cohort with a total of
19451 available general movements video snippets were randomly selected for
human clinical reasoning and computer-based analysis. We demonstrated for the
first time that pseudonymisation by face-blurring video recordings is a viable
approach. The video redaction did not affect classification accuracy for either
human assessors or computer vision methods, suggesting an adequate and
easy-to-apply solution for sharing movement video data. We call for further
explorations into efficient and privacy rule-conforming approaches for
deidentifying video data in scientific and clinical fields beyond movement
assessments. These approaches shall enable sharing and merging stand-alone
video datasets into large data pools to advance science and public health.
| no_new_dataset | 0.936923 |
2208.03543 | Chaoqiang Zhao | Chaoqiang Zhao, Youmin Zhang, Matteo Poggi, Fabio Tosi, Xianda Guo,
Zheng Zhu, Guan Huang, Yang Tang, Stefano Mattoccia | MonoViT: Self-Supervised Monocular Depth Estimation with a Vision
Transformer | Accepted by 3DV 2022 | null | 10.1109/3DV57658.2022.00077 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Self-supervised monocular depth estimation is an attractive solution that
does not require hard-to-source depth labels for training. Convolutional neural
networks (CNNs) have recently achieved great success in this task. However,
their limited receptive field constrains existing network architectures to
reason only locally, dampening the effectiveness of the self-supervised
paradigm. In the light of the recent successes achieved by Vision Transformers
(ViTs), we propose MonoViT, a brand-new framework combining the global
reasoning enabled by ViT models with the flexibility of self-supervised
monocular depth estimation. By combining plain convolutions with Transformer
blocks, our model can reason locally and globally, yielding depth prediction at
a higher level of detail and accuracy, allowing MonoViT to achieve
state-of-the-art performance on the established KITTI dataset. Moreover,
MonoViT proves its superior generalization capacities on other datasets such as
Make3D and DrivingStereo.
| [
{
"version": "v1",
"created": "Sat, 6 Aug 2022 16:54:45 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Zhao",
"Chaoqiang",
""
],
[
"Zhang",
"Youmin",
""
],
[
"Poggi",
"Matteo",
""
],
[
"Tosi",
"Fabio",
""
],
[
"Guo",
"Xianda",
""
],
[
"Zhu",
"Zheng",
""
],
[
"Huang",
"Guan",
""
],
[
"Tang",
"Yang",
""
],
[
"Mattoccia",
"Stefano",
""
]
] | TITLE: MonoViT: Self-Supervised Monocular Depth Estimation with a Vision
Transformer
ABSTRACT: Self-supervised monocular depth estimation is an attractive solution that
does not require hard-to-source depth labels for training. Convolutional neural
networks (CNNs) have recently achieved great success in this task. However,
their limited receptive field constrains existing network architectures to
reason only locally, dampening the effectiveness of the self-supervised
paradigm. In the light of the recent successes achieved by Vision Transformers
(ViTs), we propose MonoViT, a brand-new framework combining the global
reasoning enabled by ViT models with the flexibility of self-supervised
monocular depth estimation. By combining plain convolutions with Transformer
blocks, our model can reason locally and globally, yielding depth prediction at
a higher level of detail and accuracy, allowing MonoViT to achieve
state-of-the-art performance on the established KITTI dataset. Moreover,
MonoViT proves its superior generalization capacities on other datasets such as
Make3D and DrivingStereo.
| no_new_dataset | 0.946892 |
2208.08888 | Le-Anh Tran | Le-Anh Tran, Henock M. Deberneh, Truong-Dong Do, Thanh-Dat Nguyen,
My-Ha Le, Dong-Chul Park | POCS-based Clustering Algorithm | 6 pages, 4 figures, IWIS 2022 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A novel clustering technique based on the projection onto convex set (POCS)
method, called POCS-based clustering algorithm, is proposed in this paper. The
proposed POCS-based clustering algorithm exploits a parallel projection method
of POCS to find appropriate cluster prototypes in the feature space. The
algorithm considers each data point as a convex set and projects the cluster
prototypes parallelly to the member data points. The projections are convexly
combined to minimize the objective function for data clustering purpose. The
performance of the proposed POCS-based clustering algorithm is verified through
experiments on various synthetic datasets. The experimental results show that
the proposed POCS-based clustering algorithm is competitive and efficient in
terms of clustering error and execution speed when compared with other
conventional clustering methods including Fuzzy C-Means (FCM) and K-means
clustering algorithms.
| [
{
"version": "v1",
"created": "Mon, 15 Aug 2022 12:33:09 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Sep 2022 07:35:15 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 06:35:45 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Tran",
"Le-Anh",
""
],
[
"Deberneh",
"Henock M.",
""
],
[
"Do",
"Truong-Dong",
""
],
[
"Nguyen",
"Thanh-Dat",
""
],
[
"Le",
"My-Ha",
""
],
[
"Park",
"Dong-Chul",
""
]
] | TITLE: POCS-based Clustering Algorithm
ABSTRACT: A novel clustering technique based on the projection onto convex set (POCS)
method, called POCS-based clustering algorithm, is proposed in this paper. The
proposed POCS-based clustering algorithm exploits a parallel projection method
of POCS to find appropriate cluster prototypes in the feature space. The
algorithm considers each data point as a convex set and projects the cluster
prototypes parallelly to the member data points. The projections are convexly
combined to minimize the objective function for data clustering purpose. The
performance of the proposed POCS-based clustering algorithm is verified through
experiments on various synthetic datasets. The experimental results show that
the proposed POCS-based clustering algorithm is competitive and efficient in
terms of clustering error and execution speed when compared with other
conventional clustering methods including Fuzzy C-Means (FCM) and K-means
clustering algorithms.
| no_new_dataset | 0.954774 |
2209.05375 | Carlos Mastalli | Carlos Mastalli, Saroj Prasad Chhatoi, Thomas Corb\`eres, Steve
Tonneau, Sethu Vijayakumar | Inverse-Dynamics MPC via Nullspace Resolution | 20 pages, 14 figures, accepted to IEEE TRO | IEEE Transactions on Robotics, 2023 | null | null | cs.RO cs.CE cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | Optimal control (OC) using inverse dynamics provides numerical benefits such
as coarse optimization, cheaper computation of derivatives, and a high
convergence rate. However, to take advantage of these benefits in model
predictive control (MPC) for legged robots, it is crucial to handle efficiently
its large number of equality constraints. To accomplish this, we first (i)
propose a novel approach to handle equality constraints based on nullspace
parametrization. Our approach balances optimality, and both dynamics and
equality-constraint feasibility appropriately, which increases the basin of
attraction to high-quality local minima. To do so, we (ii) modify our
feasibility-driven search by incorporating a merit function. Furthermore, we
introduce (iii) a condensed formulation of inverse dynamics that considers
arbitrary actuator models. We also propose (iv) a novel MPC based on inverse
dynamics within a perceptive locomotion framework. Finally, we present (v) a
theoretical comparison of optimal control with forward and inverse dynamics and
evaluate both numerically. Our approach enables the first application of
inverse-dynamics MPC on hardware, resulting in state-of-the-art dynamic
climbing on the ANYmal robot. We benchmark it over a wide range of robotics
problems and generate agile and complex maneuvers. We show the computational
reduction of our nullspace resolution and condensed formulation (up to 47.3%).
We provide evidence of the benefits of our approach by solving coarse
optimization problems with a high convergence rate (up to 10 Hz of
discretization). Our algorithm is publicly available inside CROCODDYL.
| [
{
"version": "v1",
"created": "Mon, 12 Sep 2022 16:25:32 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jan 2023 12:21:35 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 15:56:47 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Mastalli",
"Carlos",
""
],
[
"Chhatoi",
"Saroj Prasad",
""
],
[
"Corbères",
"Thomas",
""
],
[
"Tonneau",
"Steve",
""
],
[
"Vijayakumar",
"Sethu",
""
]
] | TITLE: Inverse-Dynamics MPC via Nullspace Resolution
ABSTRACT: Optimal control (OC) using inverse dynamics provides numerical benefits such
as coarse optimization, cheaper computation of derivatives, and a high
convergence rate. However, to take advantage of these benefits in model
predictive control (MPC) for legged robots, it is crucial to handle efficiently
its large number of equality constraints. To accomplish this, we first (i)
propose a novel approach to handle equality constraints based on nullspace
parametrization. Our approach balances optimality, and both dynamics and
equality-constraint feasibility appropriately, which increases the basin of
attraction to high-quality local minima. To do so, we (ii) modify our
feasibility-driven search by incorporating a merit function. Furthermore, we
introduce (iii) a condensed formulation of inverse dynamics that considers
arbitrary actuator models. We also propose (iv) a novel MPC based on inverse
dynamics within a perceptive locomotion framework. Finally, we present (v) a
theoretical comparison of optimal control with forward and inverse dynamics and
evaluate both numerically. Our approach enables the first application of
inverse-dynamics MPC on hardware, resulting in state-of-the-art dynamic
climbing on the ANYmal robot. We benchmark it over a wide range of robotics
problems and generate agile and complex maneuvers. We show the computational
reduction of our nullspace resolution and condensed formulation (up to 47.3%).
We provide evidence of the benefits of our approach by solving coarse
optimization problems with a high convergence rate (up to 10 Hz of
discretization). Our algorithm is publicly available inside CROCODDYL.
| no_new_dataset | 0.942929 |
2209.07888 | Mathieu Gonzalez | Mathieu Gonzalez, Eric Marchand, Amine Kacete and J\'er\^ome Royan | TwistSLAM++: Fusing multiple modalities for accurate dynamic semantic
SLAM | null | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most classical SLAM systems rely on the static scene assumption, which limits
their applicability in real world scenarios. Recent SLAM frameworks have been
proposed to simultaneously track the camera and moving objects. However they
are often unable to estimate the canonical pose of the objects and exhibit a
low object tracking accuracy. To solve this problem we propose TwistSLAM++, a
semantic, dynamic, SLAM system that fuses stereo images and LiDAR information.
Using semantic information, we track potentially moving objects and associate
them to 3D object detections in LiDAR scans to obtain their pose and size.
Then, we perform registration on consecutive object scans to refine object pose
estimation. Finally, object scans are used to estimate the shape of the object
and constrain map points to lie on the estimated surface within the BA. We show
on classical benchmarks that this fusion approach based on multimodal
information improves the accuracy of object tracking.
| [
{
"version": "v1",
"created": "Fri, 16 Sep 2022 12:28:21 GMT"
},
{
"version": "v2",
"created": "Wed, 22 Mar 2023 20:20:48 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Gonzalez",
"Mathieu",
""
],
[
"Marchand",
"Eric",
""
],
[
"Kacete",
"Amine",
""
],
[
"Royan",
"Jérôme",
""
]
] | TITLE: TwistSLAM++: Fusing multiple modalities for accurate dynamic semantic
SLAM
ABSTRACT: Most classical SLAM systems rely on the static scene assumption, which limits
their applicability in real world scenarios. Recent SLAM frameworks have been
proposed to simultaneously track the camera and moving objects. However they
are often unable to estimate the canonical pose of the objects and exhibit a
low object tracking accuracy. To solve this problem we propose TwistSLAM++, a
semantic, dynamic, SLAM system that fuses stereo images and LiDAR information.
Using semantic information, we track potentially moving objects and associate
them to 3D object detections in LiDAR scans to obtain their pose and size.
Then, we perform registration on consecutive object scans to refine object pose
estimation. Finally, object scans are used to estimate the shape of the object
and constrain map points to lie on the estimated surface within the BA. We show
on classical benchmarks that this fusion approach based on multimodal
information improves the accuracy of object tracking.
| no_new_dataset | 0.950549 |
2209.08299 | Christoph Wernhard | Michael Benedikt, C\'ecilia Pradic, Christoph Wernhard | Synthesizing Nested Relational Queries from Implicit Specifications | null | null | null | null | cs.DB cs.LO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Derived datasets can be defined implicitly or explicitly. An implicit
definition (of dataset $O$ in terms of datasets $\vec{I}$) is a logical
specification involving the source data $\vec{I}$ and the interface data $O$.
It is a valid definition of $O$ in terms of $\vec{I}$, if any two models of the
specification agreeing on $\vec{I}$ agree on $O$. In contrast, an explicit
definition is a query that produces $O$ from $\vec{I}$. Variants of Beth's
theorem state that one can convert implicit definitions to explicit ones.
Further, this conversion can be done effectively given a proof witnessing
implicit definability in a suitable proof system. We prove the analogous
effective implicit-to-explicit result for nested relations: implicit
definitions, given in the natural logic for nested relations, can be
effectively converted to explicit definitions in the nested relational calculus
NRC. As a consequence, we can effectively extract rewritings of NRC queries in
terms of NRC views, given a proof witnessing that the query is determined by
the views.
| [
{
"version": "v1",
"created": "Sat, 17 Sep 2022 09:41:04 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Benedikt",
"Michael",
""
],
[
"Pradic",
"Cécilia",
""
],
[
"Wernhard",
"Christoph",
""
]
] | TITLE: Synthesizing Nested Relational Queries from Implicit Specifications
ABSTRACT: Derived datasets can be defined implicitly or explicitly. An implicit
definition (of dataset $O$ in terms of datasets $\vec{I}$) is a logical
specification involving the source data $\vec{I}$ and the interface data $O$.
It is a valid definition of $O$ in terms of $\vec{I}$, if any two models of the
specification agreeing on $\vec{I}$ agree on $O$. In contrast, an explicit
definition is a query that produces $O$ from $\vec{I}$. Variants of Beth's
theorem state that one can convert implicit definitions to explicit ones.
Further, this conversion can be done effectively given a proof witnessing
implicit definability in a suitable proof system. We prove the analogous
effective implicit-to-explicit result for nested relations: implicit
definitions, given in the natural logic for nested relations, can be
effectively converted to explicit definitions in the nested relational calculus
NRC. As a consequence, we can effectively extract rewritings of NRC queries in
terms of NRC views, given a proof witnessing that the query is determined by
the views.
| no_new_dataset | 0.934813 |
2210.00924 | Rasmus Haugaard | Rasmus Laurvig Haugaard, Thorbj{\o}rn Mosekj{\ae}r Iversen | Multi-view object pose estimation from correspondence distributions and
epipolar geometry | 7 pages, 2 figures, 1 table, ICRA 2023 | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In many automation tasks involving manipulation of rigid objects, the poses
of the objects must be acquired. Vision-based pose estimation using a single
RGB or RGB-D sensor is especially popular due to its broad applicability.
However, single-view pose estimation is inherently limited by depth ambiguity
and ambiguities imposed by various phenomena like occlusion, self-occlusion,
reflections, etc. Aggregation of information from multiple views can
potentially resolve these ambiguities, but the current state-of-the-art
multi-view pose estimation method only uses multiple views to aggregate
single-view pose estimates, and thus rely on obtaining good single-view
estimates. We present a multi-view pose estimation method which aggregates
learned 2D-3D distributions from multiple views for both the initial estimate
and optional refinement. Our method performs probabilistic sampling of 3D-3D
correspondences under epipolar constraints using learned 2D-3D correspondence
distributions which are implicitly trained to respect visual ambiguities such
as symmetry. Evaluation on the T-LESS dataset shows that our method reduces
pose estimation errors by 80-91% compared to the best single-view method, and
we present state-of-the-art results on T-LESS with four views, even compared
with methods using five and eight views.
| [
{
"version": "v1",
"created": "Mon, 3 Oct 2022 13:30:40 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 13:02:42 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Haugaard",
"Rasmus Laurvig",
""
],
[
"Iversen",
"Thorbjørn Mosekjær",
""
]
] | TITLE: Multi-view object pose estimation from correspondence distributions and
epipolar geometry
ABSTRACT: In many automation tasks involving manipulation of rigid objects, the poses
of the objects must be acquired. Vision-based pose estimation using a single
RGB or RGB-D sensor is especially popular due to its broad applicability.
However, single-view pose estimation is inherently limited by depth ambiguity
and ambiguities imposed by various phenomena like occlusion, self-occlusion,
reflections, etc. Aggregation of information from multiple views can
potentially resolve these ambiguities, but the current state-of-the-art
multi-view pose estimation method only uses multiple views to aggregate
single-view pose estimates, and thus rely on obtaining good single-view
estimates. We present a multi-view pose estimation method which aggregates
learned 2D-3D distributions from multiple views for both the initial estimate
and optional refinement. Our method performs probabilistic sampling of 3D-3D
correspondences under epipolar constraints using learned 2D-3D correspondence
distributions which are implicitly trained to respect visual ambiguities such
as symmetry. Evaluation on the T-LESS dataset shows that our method reduces
pose estimation errors by 80-91% compared to the best single-view method, and
we present state-of-the-art results on T-LESS with four views, even compared
with methods using five and eight views.
| no_new_dataset | 0.941601 |
2210.01117 | Ziming Liu | Ziming Liu, Eric J. Michaud, Max Tegmark | Omnigrok: Grokking Beyond Algorithmic Data | null | null | null | null | cs.LG cs.AI physics.data-an stat.ME stat.ML | http://creativecommons.org/licenses/by/4.0/ | Grokking, the unusual phenomenon for algorithmic datasets where
generalization happens long after overfitting the training data, has remained
elusive. We aim to understand grokking by analyzing the loss landscapes of
neural networks, identifying the mismatch between training and test losses as
the cause for grokking. We refer to this as the "LU mechanism" because training
and test losses (against model weight norm) typically resemble "L" and "U",
respectively. This simple mechanism can nicely explain many aspects of
grokking: data size dependence, weight decay dependence, the emergence of
representations, etc. Guided by the intuitive picture, we are able to induce
grokking on tasks involving images, language and molecules. In the reverse
direction, we are able to eliminate grokking for algorithmic datasets. We
attribute the dramatic nature of grokking for algorithmic datasets to
representation learning.
| [
{
"version": "v1",
"created": "Mon, 3 Oct 2022 17:58:04 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 13:42:27 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Liu",
"Ziming",
""
],
[
"Michaud",
"Eric J.",
""
],
[
"Tegmark",
"Max",
""
]
] | TITLE: Omnigrok: Grokking Beyond Algorithmic Data
ABSTRACT: Grokking, the unusual phenomenon for algorithmic datasets where
generalization happens long after overfitting the training data, has remained
elusive. We aim to understand grokking by analyzing the loss landscapes of
neural networks, identifying the mismatch between training and test losses as
the cause for grokking. We refer to this as the "LU mechanism" because training
and test losses (against model weight norm) typically resemble "L" and "U",
respectively. This simple mechanism can nicely explain many aspects of
grokking: data size dependence, weight decay dependence, the emergence of
representations, etc. Guided by the intuitive picture, we are able to induce
grokking on tasks involving images, language and molecules. In the reverse
direction, we are able to eliminate grokking for algorithmic datasets. We
attribute the dramatic nature of grokking for algorithmic datasets to
representation learning.
| no_new_dataset | 0.947866 |
2210.03591 | Wenbin Li | Wenbin Li, Zhichen Fan, Jing Huo, Yang Gao | Modeling Inter-Class and Intra-Class Constraints in Novel Class
Discovery | Accepted to CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Novel class discovery (NCD) aims at learning a model that transfers the
common knowledge from a class-disjoint labelled dataset to another unlabelled
dataset and discovers new classes (clusters) within it. Many methods, as well
as elaborate training pipelines and appropriate objectives, have been proposed
and considerably boosted performance on NCD tasks. Despite all this, we find
that the existing methods do not sufficiently take advantage of the essence of
the NCD setting. To this end, in this paper, we propose to model both
inter-class and intra-class constraints in NCD based on the symmetric
Kullback-Leibler divergence (sKLD). Specifically, we propose an inter-class
sKLD constraint to effectively exploit the disjoint relationship between
labelled and unlabelled classes, enforcing the separability for different
classes in the embedding space. In addition, we present an intra-class sKLD
constraint to explicitly constrain the intra-relationship between a sample and
its augmentations and ensure the stability of the training process at the same
time. We conduct extensive experiments on the popular CIFAR10, CIFAR100 and
ImageNet benchmarks and successfully demonstrate that our method can establish
a new state of the art and can achieve significant performance improvements,
e.g., 3.5%/3.7% clustering accuracy improvements on CIFAR100-50 dataset split
under the task-aware/-agnostic evaluation protocol, over previous
state-of-the-art methods. Code is available at
https://github.com/FanZhichen/NCD-IIC.
| [
{
"version": "v1",
"created": "Fri, 7 Oct 2022 14:46:32 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Oct 2022 07:17:40 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 13:15:27 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Li",
"Wenbin",
""
],
[
"Fan",
"Zhichen",
""
],
[
"Huo",
"Jing",
""
],
[
"Gao",
"Yang",
""
]
] | TITLE: Modeling Inter-Class and Intra-Class Constraints in Novel Class
Discovery
ABSTRACT: Novel class discovery (NCD) aims at learning a model that transfers the
common knowledge from a class-disjoint labelled dataset to another unlabelled
dataset and discovers new classes (clusters) within it. Many methods, as well
as elaborate training pipelines and appropriate objectives, have been proposed
and considerably boosted performance on NCD tasks. Despite all this, we find
that the existing methods do not sufficiently take advantage of the essence of
the NCD setting. To this end, in this paper, we propose to model both
inter-class and intra-class constraints in NCD based on the symmetric
Kullback-Leibler divergence (sKLD). Specifically, we propose an inter-class
sKLD constraint to effectively exploit the disjoint relationship between
labelled and unlabelled classes, enforcing the separability for different
classes in the embedding space. In addition, we present an intra-class sKLD
constraint to explicitly constrain the intra-relationship between a sample and
its augmentations and ensure the stability of the training process at the same
time. We conduct extensive experiments on the popular CIFAR10, CIFAR100 and
ImageNet benchmarks and successfully demonstrate that our method can establish
a new state of the art and can achieve significant performance improvements,
e.g., 3.5%/3.7% clustering accuracy improvements on CIFAR100-50 dataset split
under the task-aware/-agnostic evaluation protocol, over previous
state-of-the-art methods. Code is available at
https://github.com/FanZhichen/NCD-IIC.
| no_new_dataset | 0.943452 |
2211.01367 | Fangyun Wei | Yutong Chen, Ronglai Zuo, Fangyun Wei, Yu Wu, Shujie Liu, Brian Mak | Two-Stream Network for Sign Language Recognition and Translation | Accepted by NeurIPS 2022. Code and models are available at:
https://github.com/FangyunWei/SLRT | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sign languages are visual languages using manual articulations and non-manual
elements to convey information. For sign language recognition and translation,
the majority of existing approaches directly encode RGB videos into hidden
representations. RGB videos, however, are raw signals with substantial visual
redundancy, leading the encoder to overlook the key information for sign
language understanding. To mitigate this problem and better incorporate domain
knowledge, such as handshape and body movement, we introduce a dual visual
encoder containing two separate streams to model both the raw videos and the
keypoint sequences generated by an off-the-shelf keypoint estimator. To make
the two streams interact with each other, we explore a variety of techniques,
including bidirectional lateral connection, sign pyramid network with auxiliary
supervision, and frame-level self-distillation. The resulting model is called
TwoStream-SLR, which is competent for sign language recognition (SLR).
TwoStream-SLR is extended to a sign language translation (SLT) model,
TwoStream-SLT, by simply attaching an extra translation network.
Experimentally, our TwoStream-SLR and TwoStream-SLT achieve state-of-the-art
performance on SLR and SLT tasks across a series of datasets including
Phoenix-2014, Phoenix-2014T, and CSL-Daily. Code and models are available at:
https://github.com/FangyunWei/SLRT.
| [
{
"version": "v1",
"created": "Wed, 2 Nov 2022 17:59:58 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 02:49:35 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Chen",
"Yutong",
""
],
[
"Zuo",
"Ronglai",
""
],
[
"Wei",
"Fangyun",
""
],
[
"Wu",
"Yu",
""
],
[
"Liu",
"Shujie",
""
],
[
"Mak",
"Brian",
""
]
] | TITLE: Two-Stream Network for Sign Language Recognition and Translation
ABSTRACT: Sign languages are visual languages using manual articulations and non-manual
elements to convey information. For sign language recognition and translation,
the majority of existing approaches directly encode RGB videos into hidden
representations. RGB videos, however, are raw signals with substantial visual
redundancy, leading the encoder to overlook the key information for sign
language understanding. To mitigate this problem and better incorporate domain
knowledge, such as handshape and body movement, we introduce a dual visual
encoder containing two separate streams to model both the raw videos and the
keypoint sequences generated by an off-the-shelf keypoint estimator. To make
the two streams interact with each other, we explore a variety of techniques,
including bidirectional lateral connection, sign pyramid network with auxiliary
supervision, and frame-level self-distillation. The resulting model is called
TwoStream-SLR, which is competent for sign language recognition (SLR).
TwoStream-SLR is extended to a sign language translation (SLT) model,
TwoStream-SLT, by simply attaching an extra translation network.
Experimentally, our TwoStream-SLR and TwoStream-SLT achieve state-of-the-art
performance on SLR and SLT tasks across a series of datasets including
Phoenix-2014, Phoenix-2014T, and CSL-Daily. Code and models are available at:
https://github.com/FangyunWei/SLRT.
| no_new_dataset | 0.942876 |
2211.03456 | Xin Jin | Xin Jin, Longhai Wu, Jie Chen, Youxin Chen, Jayoon Koo, Cheul-hee Hahm | A Unified Pyramid Recurrent Network for Video Frame Interpolation | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Flow-guided synthesis provides a common framework for frame interpolation,
where optical flow is estimated to guide the synthesis of intermediate frames
between consecutive inputs. In this paper, we present UPR-Net, a novel Unified
Pyramid Recurrent Network for frame interpolation. Cast in a flexible pyramid
framework, UPR-Net exploits lightweight recurrent modules for both
bi-directional flow estimation and intermediate frame synthesis. At each
pyramid level, it leverages estimated bi-directional flow to generate
forward-warped representations for frame synthesis; across pyramid levels, it
enables iterative refinement for both optical flow and intermediate frame. In
particular, we show that our iterative synthesis strategy can significantly
improve the robustness of frame interpolation on large motion cases. Despite
being extremely lightweight (1.7M parameters), our base version of UPR-Net
achieves excellent performance on a large range of benchmarks. Code and trained
models of our UPR-Net series are available at:
https://github.com/srcn-ivl/UPR-Net.
| [
{
"version": "v1",
"created": "Mon, 7 Nov 2022 11:12:31 GMT"
},
{
"version": "v2",
"created": "Thu, 23 Mar 2023 04:14:45 GMT"
}
] | 2023-03-24T00:00:00 | [
[
"Jin",
"Xin",
""
],
[
"Wu",
"Longhai",
""
],
[
"Chen",
"Jie",
""
],
[
"Chen",
"Youxin",
""
],
[
"Koo",
"Jayoon",
""
],
[
"Hahm",
"Cheul-hee",
""
]
] | TITLE: A Unified Pyramid Recurrent Network for Video Frame Interpolation
ABSTRACT: Flow-guided synthesis provides a common framework for frame interpolation,
where optical flow is estimated to guide the synthesis of intermediate frames
between consecutive inputs. In this paper, we present UPR-Net, a novel Unified
Pyramid Recurrent Network for frame interpolation. Cast in a flexible pyramid
framework, UPR-Net exploits lightweight recurrent modules for both
bi-directional flow estimation and intermediate frame synthesis. At each
pyramid level, it leverages estimated bi-directional flow to generate
forward-warped representations for frame synthesis; across pyramid levels, it
enables iterative refinement for both optical flow and intermediate frame. In
particular, we show that our iterative synthesis strategy can significantly
improve the robustness of frame interpolation on large motion cases. Despite
being extremely lightweight (1.7M parameters), our base version of UPR-Net
achieves excellent performance on a large range of benchmarks. Code and trained
models of our UPR-Net series are available at:
https://github.com/srcn-ivl/UPR-Net.
| no_new_dataset | 0.948394 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.