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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2303.14292 | Paula Maddigan | Paula Maddigan and Teo Susnjak | Chat2VIS: Fine-Tuning Data Visualisations using Multilingual Natural
Language Text and Pre-Trained Large Language Models | null | null | null | null | cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The explosion of data in recent years is driving individuals to leverage
technology to generate insights. Traditional tools bring heavy learning
overheads and the requirement for understanding complex charting techniques.
Such barriers can hinder those who may benefit from harnessing data for
informed decision making. The emerging field of generating data visualisations
from natural language text (NL2VIS) addresses this issue. This study showcases
Chat2VIS, a state-of-the-art NL2VIS solution. It capitalises on the latest in
AI technology with the upsurge in pre-trained large language models (LLMs) such
as GPT-3, Codex, and ChatGPT. Furthermore, the rise in natural language
interfaces (NLI) and chatbots is taking centre stage. This work illustrates how
Chat2VIS leverages similar techniques to fine-tune data visualisation
components beyond that demonstrated in previous approaches. In addition, this
paper presents the flexibility of Chat2VIS to comprehend multilingual natural
language requests. No other NL2VIS system has demonstrated this unique talent.
In concluding, this research provides quantitative benchmarking evaluations to
contribute to the paucity of NL2VIS standards.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 22:21:32 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Maddigan",
"Paula",
""
],
[
"Susnjak",
"Teo",
""
]
] | TITLE: Chat2VIS: Fine-Tuning Data Visualisations using Multilingual Natural
Language Text and Pre-Trained Large Language Models
ABSTRACT: The explosion of data in recent years is driving individuals to leverage
technology to generate insights. Traditional tools bring heavy learning
overheads and the requirement for understanding complex charting techniques.
Such barriers can hinder those who may benefit from harnessing data for
informed decision making. The emerging field of generating data visualisations
from natural language text (NL2VIS) addresses this issue. This study showcases
Chat2VIS, a state-of-the-art NL2VIS solution. It capitalises on the latest in
AI technology with the upsurge in pre-trained large language models (LLMs) such
as GPT-3, Codex, and ChatGPT. Furthermore, the rise in natural language
interfaces (NLI) and chatbots is taking centre stage. This work illustrates how
Chat2VIS leverages similar techniques to fine-tune data visualisation
components beyond that demonstrated in previous approaches. In addition, this
paper presents the flexibility of Chat2VIS to comprehend multilingual natural
language requests. No other NL2VIS system has demonstrated this unique talent.
In concluding, this research provides quantitative benchmarking evaluations to
contribute to the paucity of NL2VIS standards.
| no_new_dataset | 0.944382 |
2303.14322 | Usman Nazir | U. Nazir, W. Islam, M. Taj | Spatio-Temporal driven Attention Graph Neural Network with Block
Adjacency matrix (STAG-NN-BA) | null | KDD 2023 | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite the recent advances in deep neural networks, standard convolutional
kernels limit the applications of these networks to the Euclidean domain only.
Considering the geodesic nature of the measurement of the earth's surface,
remote sensing is one such area that can benefit from non-Euclidean and
spherical domains. For this purpose, we propose a novel Graph Neural Network
architecture for spatial and spatio-temporal classification using satellite
imagery. We propose a hybrid attention method to learn the relative importance
of irregular neighbors in remote sensing data. Instead of classifying each
pixel, we propose a method based on Simple Linear Iterative Clustering (SLIC)
image segmentation and Graph Attention GAT. The superpixels obtained from SLIC
become the nodes of our Graph Convolution Network (GCN). We then construct a
region adjacency graph (RAG) where each superpixel is connected to every other
adjacent superpixel in the image, enabling information to propagate globally.
Finally, we propose a Spatially driven Attention Graph Neural Network (SAG-NN)
to classify each RAG. We also propose an extension to our SAG-NN for
spatio-temporal data. Unlike regular grids of pixels in images, superpixels are
irregular in nature and cannot be used to create spatio-temporal graphs. We
introduce temporal bias by combining unconnected RAGs from each image into one
supergraph. This is achieved by introducing block adjacency matrices resulting
in novel Spatio-Temporal driven Attention Graph Neural Network with Block
Adjacency matrix (STAG-NN-BA). We evaluate our proposed methods on two remote
sensing datasets namely Asia14 and C2D2. In comparison with both non-graph and
graph-based approaches our SAG-NN and STAG-NN-BA achieved superior accuracy on
all the datasets while incurring less computation cost. The code and dataset
will be made public via our GitHub repository.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 01:26:50 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Nazir",
"U.",
""
],
[
"Islam",
"W.",
""
],
[
"Taj",
"M.",
""
]
] | TITLE: Spatio-Temporal driven Attention Graph Neural Network with Block
Adjacency matrix (STAG-NN-BA)
ABSTRACT: Despite the recent advances in deep neural networks, standard convolutional
kernels limit the applications of these networks to the Euclidean domain only.
Considering the geodesic nature of the measurement of the earth's surface,
remote sensing is one such area that can benefit from non-Euclidean and
spherical domains. For this purpose, we propose a novel Graph Neural Network
architecture for spatial and spatio-temporal classification using satellite
imagery. We propose a hybrid attention method to learn the relative importance
of irregular neighbors in remote sensing data. Instead of classifying each
pixel, we propose a method based on Simple Linear Iterative Clustering (SLIC)
image segmentation and Graph Attention GAT. The superpixels obtained from SLIC
become the nodes of our Graph Convolution Network (GCN). We then construct a
region adjacency graph (RAG) where each superpixel is connected to every other
adjacent superpixel in the image, enabling information to propagate globally.
Finally, we propose a Spatially driven Attention Graph Neural Network (SAG-NN)
to classify each RAG. We also propose an extension to our SAG-NN for
spatio-temporal data. Unlike regular grids of pixels in images, superpixels are
irregular in nature and cannot be used to create spatio-temporal graphs. We
introduce temporal bias by combining unconnected RAGs from each image into one
supergraph. This is achieved by introducing block adjacency matrices resulting
in novel Spatio-Temporal driven Attention Graph Neural Network with Block
Adjacency matrix (STAG-NN-BA). We evaluate our proposed methods on two remote
sensing datasets namely Asia14 and C2D2. In comparison with both non-graph and
graph-based approaches our SAG-NN and STAG-NN-BA achieved superior accuracy on
all the datasets while incurring less computation cost. The code and dataset
will be made public via our GitHub repository.
| no_new_dataset | 0.953579 |
2303.14325 | Xukun Zhou | Xukun Zhou, Jiwei Li, Tianwei Zhang, Lingjuan Lyu, Muqiao Yang, Jun He | Backdoor Attacks with Input-unique Triggers in NLP | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Backdoor attack aims at inducing neural models to make incorrect predictions
for poison data while keeping predictions on the clean dataset unchanged, which
creates a considerable threat to current natural language processing (NLP)
systems. Existing backdoor attacking systems face two severe issues:firstly,
most backdoor triggers follow a uniform and usually input-independent pattern,
e.g., insertion of specific trigger words, synonym replacement. This
significantly hinders the stealthiness of the attacking model, leading the
trained backdoor model being easily identified as malicious by model probes.
Secondly, trigger-inserted poisoned sentences are usually disfluent,
ungrammatical, or even change the semantic meaning from the original sentence,
making them being easily filtered in the pre-processing stage. To resolve these
two issues, in this paper, we propose an input-unique backdoor attack(NURA),
where we generate backdoor triggers unique to inputs. IDBA generates
context-related triggers by continuing writing the input with a language model
like GPT2. The generated sentence is used as the backdoor trigger. This
strategy not only creates input-unique backdoor triggers, but also preserves
the semantics of the original input, simultaneously resolving the two issues
above. Experimental results show that the IDBA attack is effective for attack
and difficult to defend: it achieves high attack success rate across all the
widely applied benchmarks, while is immune to existing defending methods. In
addition, it is able to generate fluent, grammatical, and diverse backdoor
inputs, which can hardly be recognized through human inspection.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 01:41:54 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhou",
"Xukun",
""
],
[
"Li",
"Jiwei",
""
],
[
"Zhang",
"Tianwei",
""
],
[
"Lyu",
"Lingjuan",
""
],
[
"Yang",
"Muqiao",
""
],
[
"He",
"Jun",
""
]
] | TITLE: Backdoor Attacks with Input-unique Triggers in NLP
ABSTRACT: Backdoor attack aims at inducing neural models to make incorrect predictions
for poison data while keeping predictions on the clean dataset unchanged, which
creates a considerable threat to current natural language processing (NLP)
systems. Existing backdoor attacking systems face two severe issues:firstly,
most backdoor triggers follow a uniform and usually input-independent pattern,
e.g., insertion of specific trigger words, synonym replacement. This
significantly hinders the stealthiness of the attacking model, leading the
trained backdoor model being easily identified as malicious by model probes.
Secondly, trigger-inserted poisoned sentences are usually disfluent,
ungrammatical, or even change the semantic meaning from the original sentence,
making them being easily filtered in the pre-processing stage. To resolve these
two issues, in this paper, we propose an input-unique backdoor attack(NURA),
where we generate backdoor triggers unique to inputs. IDBA generates
context-related triggers by continuing writing the input with a language model
like GPT2. The generated sentence is used as the backdoor trigger. This
strategy not only creates input-unique backdoor triggers, but also preserves
the semantics of the original input, simultaneously resolving the two issues
above. Experimental results show that the IDBA attack is effective for attack
and difficult to defend: it achieves high attack success rate across all the
widely applied benchmarks, while is immune to existing defending methods. In
addition, it is able to generate fluent, grammatical, and diverse backdoor
inputs, which can hardly be recognized through human inspection.
| no_new_dataset | 0.946794 |
2303.14341 | Yifu Ding | Yifu Ding, Haotong Qin, Qinghua Yan, Zhenhua Chai, Junjie Liu, Xiaolin
Wei, Xianglong Liu | Towards Accurate Post-Training Quantization for Vision Transformer | 9 pages, 5 figures, accepted by ACM Multimedia 2022 | null | 10.1145/3503161.3547826 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision transformer emerges as a potential architecture for vision tasks.
However, the intense computation and non-negligible delay hinder its
application in the real world. As a widespread model compression technique,
existing post-training quantization methods still cause severe performance
drops. We find the main reasons lie in (1) the existing calibration metric is
inaccurate in measuring the quantization influence for extremely low-bit
representation, and (2) the existing quantization paradigm is unfriendly to the
power-law distribution of Softmax. Based on these observations, we propose a
novel Accurate Post-training Quantization framework for Vision Transformer,
namely APQ-ViT. We first present a unified Bottom-elimination Blockwise
Calibration scheme to optimize the calibration metric to perceive the overall
quantization disturbance in a blockwise manner and prioritize the crucial
quantization errors that influence more on the final output. Then, we design a
Matthew-effect Preserving Quantization for Softmax to maintain the power-law
character and keep the function of the attention mechanism. Comprehensive
experiments on large-scale classification and detection datasets demonstrate
that our APQ-ViT surpasses the existing post-training quantization methods by
convincing margins, especially in lower bit-width settings (e.g., averagely up
to 5.17% improvement for classification and 24.43% for detection on W4A4). We
also highlight that APQ-ViT enjoys versatility and works well on diverse
transformer variants.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 03:05:26 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ding",
"Yifu",
""
],
[
"Qin",
"Haotong",
""
],
[
"Yan",
"Qinghua",
""
],
[
"Chai",
"Zhenhua",
""
],
[
"Liu",
"Junjie",
""
],
[
"Wei",
"Xiaolin",
""
],
[
"Liu",
"Xianglong",
""
]
] | TITLE: Towards Accurate Post-Training Quantization for Vision Transformer
ABSTRACT: Vision transformer emerges as a potential architecture for vision tasks.
However, the intense computation and non-negligible delay hinder its
application in the real world. As a widespread model compression technique,
existing post-training quantization methods still cause severe performance
drops. We find the main reasons lie in (1) the existing calibration metric is
inaccurate in measuring the quantization influence for extremely low-bit
representation, and (2) the existing quantization paradigm is unfriendly to the
power-law distribution of Softmax. Based on these observations, we propose a
novel Accurate Post-training Quantization framework for Vision Transformer,
namely APQ-ViT. We first present a unified Bottom-elimination Blockwise
Calibration scheme to optimize the calibration metric to perceive the overall
quantization disturbance in a blockwise manner and prioritize the crucial
quantization errors that influence more on the final output. Then, we design a
Matthew-effect Preserving Quantization for Softmax to maintain the power-law
character and keep the function of the attention mechanism. Comprehensive
experiments on large-scale classification and detection datasets demonstrate
that our APQ-ViT surpasses the existing post-training quantization methods by
convincing margins, especially in lower bit-width settings (e.g., averagely up
to 5.17% improvement for classification and 24.43% for detection on W4A4). We
also highlight that APQ-ViT enjoys versatility and works well on diverse
transformer variants.
| no_new_dataset | 0.943608 |
2303.14347 | Yu Jiang | Ertai Liu, Josephine Monica, Kaitlin Gold, Lance Cadle-Davidson, David
Combs, Yu Jiang | Vision-based Vineyard Navigation Solution with Automatic Annotation | Submission to IROS 2023 | null | null | null | cs.RO cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Autonomous navigation is the key to achieving the full automation of
agricultural research and production management (e.g., disease management and
yield prediction) using agricultural robots. In this paper, we introduced a
vision-based autonomous navigation framework for agriculture robots in
trellised cropping systems such as vineyards. To achieve this, we proposed a
novel learning-based method to estimate the path traversibility heatmap
directly from an RGB-D image and subsequently convert the heatmap to a
preferred traversal path. An automatic annotation pipeline was developed to
form a training dataset by projecting RTK GPS paths collected during the first
setup in a vineyard in corresponding RGB-D images as ground-truth path
annotations, allowing a fast model training and fine-tuning without costly
human annotation. The trained path detection model was used to develop a full
navigation framework consisting of row tracking and row switching modules,
enabling a robot to traverse within a crop row and transit between crop rows to
cover an entire vineyard autonomously. Extensive field trials were conducted in
three different vineyards to demonstrate that the developed path detection
model and navigation framework provided a cost-effective, accurate, and robust
autonomous navigation solution in the vineyard and could be generalized to
unseen vineyards with stable performance.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 03:37:17 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Liu",
"Ertai",
""
],
[
"Monica",
"Josephine",
""
],
[
"Gold",
"Kaitlin",
""
],
[
"Cadle-Davidson",
"Lance",
""
],
[
"Combs",
"David",
""
],
[
"Jiang",
"Yu",
""
]
] | TITLE: Vision-based Vineyard Navigation Solution with Automatic Annotation
ABSTRACT: Autonomous navigation is the key to achieving the full automation of
agricultural research and production management (e.g., disease management and
yield prediction) using agricultural robots. In this paper, we introduced a
vision-based autonomous navigation framework for agriculture robots in
trellised cropping systems such as vineyards. To achieve this, we proposed a
novel learning-based method to estimate the path traversibility heatmap
directly from an RGB-D image and subsequently convert the heatmap to a
preferred traversal path. An automatic annotation pipeline was developed to
form a training dataset by projecting RTK GPS paths collected during the first
setup in a vineyard in corresponding RGB-D images as ground-truth path
annotations, allowing a fast model training and fine-tuning without costly
human annotation. The trained path detection model was used to develop a full
navigation framework consisting of row tracking and row switching modules,
enabling a robot to traverse within a crop row and transit between crop rows to
cover an entire vineyard autonomously. Extensive field trials were conducted in
three different vineyards to demonstrate that the developed path detection
model and navigation framework provided a cost-effective, accurate, and robust
autonomous navigation solution in the vineyard and could be generalized to
unseen vineyards with stable performance.
| new_dataset | 0.969091 |
2303.14348 | Yonggang Qi | Fengyin Lin, Mingkang Li, Da Li, Timothy Hospedales, Yi-Zhe Song,
Yonggang Qi | Zero-Shot Everything Sketch-Based Image Retrieval, and in Explainable
Style | CVPR 2023 (Highlight) | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This paper studies the problem of zero-short sketch-based image retrieval
(ZS-SBIR), however with two significant differentiators to prior art (i) we
tackle all variants (inter-category, intra-category, and cross datasets) of
ZS-SBIR with just one network (``everything''), and (ii) we would really like
to understand how this sketch-photo matching operates (``explainable''). Our
key innovation lies with the realization that such a cross-modal matching
problem could be reduced to comparisons of groups of key local patches -- akin
to the seasoned ``bag-of-words'' paradigm. Just with this change, we are able
to achieve both of the aforementioned goals, with the added benefit of no
longer requiring external semantic knowledge. Technically, ours is a
transformer-based cross-modal network, with three novel components (i) a
self-attention module with a learnable tokenizer to produce visual tokens that
correspond to the most informative local regions, (ii) a cross-attention module
to compute local correspondences between the visual tokens across two
modalities, and finally (iii) a kernel-based relation network to assemble local
putative matches and produce an overall similarity metric for a sketch-photo
pair. Experiments show ours indeed delivers superior performances across all
ZS-SBIR settings. The all important explainable goal is elegantly achieved by
visualizing cross-modal token correspondences, and for the first time, via
sketch to photo synthesis by universal replacement of all matched photo
patches. Code and model are available at
\url{https://github.com/buptLinfy/ZSE-SBIR}.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 03:52:32 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Lin",
"Fengyin",
""
],
[
"Li",
"Mingkang",
""
],
[
"Li",
"Da",
""
],
[
"Hospedales",
"Timothy",
""
],
[
"Song",
"Yi-Zhe",
""
],
[
"Qi",
"Yonggang",
""
]
] | TITLE: Zero-Shot Everything Sketch-Based Image Retrieval, and in Explainable
Style
ABSTRACT: This paper studies the problem of zero-short sketch-based image retrieval
(ZS-SBIR), however with two significant differentiators to prior art (i) we
tackle all variants (inter-category, intra-category, and cross datasets) of
ZS-SBIR with just one network (``everything''), and (ii) we would really like
to understand how this sketch-photo matching operates (``explainable''). Our
key innovation lies with the realization that such a cross-modal matching
problem could be reduced to comparisons of groups of key local patches -- akin
to the seasoned ``bag-of-words'' paradigm. Just with this change, we are able
to achieve both of the aforementioned goals, with the added benefit of no
longer requiring external semantic knowledge. Technically, ours is a
transformer-based cross-modal network, with three novel components (i) a
self-attention module with a learnable tokenizer to produce visual tokens that
correspond to the most informative local regions, (ii) a cross-attention module
to compute local correspondences between the visual tokens across two
modalities, and finally (iii) a kernel-based relation network to assemble local
putative matches and produce an overall similarity metric for a sketch-photo
pair. Experiments show ours indeed delivers superior performances across all
ZS-SBIR settings. The all important explainable goal is elegantly achieved by
visualizing cross-modal token correspondences, and for the first time, via
sketch to photo synthesis by universal replacement of all matched photo
patches. Code and model are available at
\url{https://github.com/buptLinfy/ZSE-SBIR}.
| no_new_dataset | 0.951188 |
2303.14351 | Li-Hsiang Shen | Li-Hsiang Shen, Yun Ho, Kai-Ten Feng, Lie-Liang Yang, Sau-Hsuan Wu,
Jen-Ming Wu | Hierarchical Multi-Agent Multi-Armed Bandit for Resource Allocation in
Multi-LEO Satellite Constellation Networks | null | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Low Earth orbit (LEO) satellite constellation is capable of providing global
coverage area with high-rate services in the next sixth-generation (6G)
non-terrestrial network (NTN). Due to limited onboard resources of operating
power, beams, and channels, resilient and efficient resource management has
become compellingly imperative under complex interference cases. However,
different from conventional terrestrial base stations, LEO is deployed at
considerable height and under high mobility, inducing substantially long delay
and interference during transmission. As a result, acquiring the accurate
channel state information between LEOs and ground users is challenging.
Therefore, we construct a framework with a two-way transmission under unknown
channel information and no data collected at long-delay ground gateway. In this
paper, we propose hierarchical multi-agent multi-armed bandit resource
allocation for LEO constellation (mmRAL) by appropriately assigning available
radio resources. LEOs are considered as collaborative multiple macro-agents
attempting unknown trials of various actions of micro-agents of respective
resources, asymptotically achieving suitable allocation with only throughput
information. In simulations, we evaluate mmRAL in various cases of LEO
deployment, serving numbers of users and LEOs, hardware cost and outage
probability. Benefited by efficient and resilient allocation, the proposed
mmRAL system is capable of operating in homogeneous or heterogeneous orbital
planes or constellations, achieving the highest throughput performance compared
to the existing benchmarks in open literature.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 04:22:07 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Shen",
"Li-Hsiang",
""
],
[
"Ho",
"Yun",
""
],
[
"Feng",
"Kai-Ten",
""
],
[
"Yang",
"Lie-Liang",
""
],
[
"Wu",
"Sau-Hsuan",
""
],
[
"Wu",
"Jen-Ming",
""
]
] | TITLE: Hierarchical Multi-Agent Multi-Armed Bandit for Resource Allocation in
Multi-LEO Satellite Constellation Networks
ABSTRACT: Low Earth orbit (LEO) satellite constellation is capable of providing global
coverage area with high-rate services in the next sixth-generation (6G)
non-terrestrial network (NTN). Due to limited onboard resources of operating
power, beams, and channels, resilient and efficient resource management has
become compellingly imperative under complex interference cases. However,
different from conventional terrestrial base stations, LEO is deployed at
considerable height and under high mobility, inducing substantially long delay
and interference during transmission. As a result, acquiring the accurate
channel state information between LEOs and ground users is challenging.
Therefore, we construct a framework with a two-way transmission under unknown
channel information and no data collected at long-delay ground gateway. In this
paper, we propose hierarchical multi-agent multi-armed bandit resource
allocation for LEO constellation (mmRAL) by appropriately assigning available
radio resources. LEOs are considered as collaborative multiple macro-agents
attempting unknown trials of various actions of micro-agents of respective
resources, asymptotically achieving suitable allocation with only throughput
information. In simulations, we evaluate mmRAL in various cases of LEO
deployment, serving numbers of users and LEOs, hardware cost and outage
probability. Benefited by efficient and resilient allocation, the proposed
mmRAL system is capable of operating in homogeneous or heterogeneous orbital
planes or constellations, achieving the highest throughput performance compared
to the existing benchmarks in open literature.
| no_new_dataset | 0.950595 |
2303.14358 | Vincent S. Tseng | Ying-Chen Lin, Vincent S. Tseng | Multi-view knowledge distillation transformer for human action
recognition | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, Transformer-based methods have been utilized to improve the
performance of human action recognition. However, most of these studies assume
that multi-view data is complete, which may not always be the case in
real-world scenarios. Therefore, this paper presents a novel Multi-view
Knowledge Distillation Transformer (MKDT) framework that consists of a teacher
network and a student network. This framework aims to handle incomplete human
action problems in real-world applications. Specifically, the multi-view
knowledge distillation transformer uses a hierarchical vision transformer with
shifted windows to capture more spatial-temporal information. Experimental
results demonstrate that our framework outperforms the CNN-based method on
three public datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 04:47:31 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Lin",
"Ying-Chen",
""
],
[
"Tseng",
"Vincent S.",
""
]
] | TITLE: Multi-view knowledge distillation transformer for human action
recognition
ABSTRACT: Recently, Transformer-based methods have been utilized to improve the
performance of human action recognition. However, most of these studies assume
that multi-view data is complete, which may not always be the case in
real-world scenarios. Therefore, this paper presents a novel Multi-view
Knowledge Distillation Transformer (MKDT) framework that consists of a teacher
network and a student network. This framework aims to handle incomplete human
action problems in real-world applications. Specifically, the multi-view
knowledge distillation transformer uses a hierarchical vision transformer with
shifted windows to capture more spatial-temporal information. Experimental
results demonstrate that our framework outperforms the CNN-based method on
three public datasets.
| no_new_dataset | 0.944536 |
2303.14360 | Xu Zheng | Xu Zheng, Jinjing Zhu, Yexin Liu, Zidong Cao, Chong Fu, Lin Wang | Both Style and Distortion Matter: Dual-Path Unsupervised Domain
Adaptation for Panoramic Semantic Segmentation | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The ability of scene understanding has sparked active research for panoramic
image semantic segmentation. However, the performance is hampered by distortion
of the equirectangular projection (ERP) and a lack of pixel-wise annotations.
For this reason, some works treat the ERP and pinhole images equally and
transfer knowledge from the pinhole to ERP images via unsupervised domain
adaptation (UDA). However, they fail to handle the domain gaps caused by: 1)
the inherent differences between camera sensors and captured scenes; 2) the
distinct image formats (e.g., ERP and pinhole images). In this paper, we
propose a novel yet flexible dual-path UDA framework, DPPASS, taking ERP and
tangent projection (TP) images as inputs. To reduce the domain gaps, we propose
cross-projection and intra-projection training. The cross-projection training
includes tangent-wise feature contrastive training and prediction consistency
training. That is, the former formulates the features with the same projection
locations as positive examples and vice versa, for the models' awareness of
distortion, while the latter ensures the consistency of cross-model predictions
between the ERP and TP. Moreover, adversarial intra-projection training is
proposed to reduce the inherent gap, between the features of the pinhole images
and those of the ERP and TP images, respectively. Importantly, the TP path can
be freely removed after training, leading to no additional inference cost.
Extensive experiments on two benchmarks show that our DPPASS achieves +1.06$\%$
mIoU increment than the state-of-the-art approaches.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 04:57:45 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zheng",
"Xu",
""
],
[
"Zhu",
"Jinjing",
""
],
[
"Liu",
"Yexin",
""
],
[
"Cao",
"Zidong",
""
],
[
"Fu",
"Chong",
""
],
[
"Wang",
"Lin",
""
]
] | TITLE: Both Style and Distortion Matter: Dual-Path Unsupervised Domain
Adaptation for Panoramic Semantic Segmentation
ABSTRACT: The ability of scene understanding has sparked active research for panoramic
image semantic segmentation. However, the performance is hampered by distortion
of the equirectangular projection (ERP) and a lack of pixel-wise annotations.
For this reason, some works treat the ERP and pinhole images equally and
transfer knowledge from the pinhole to ERP images via unsupervised domain
adaptation (UDA). However, they fail to handle the domain gaps caused by: 1)
the inherent differences between camera sensors and captured scenes; 2) the
distinct image formats (e.g., ERP and pinhole images). In this paper, we
propose a novel yet flexible dual-path UDA framework, DPPASS, taking ERP and
tangent projection (TP) images as inputs. To reduce the domain gaps, we propose
cross-projection and intra-projection training. The cross-projection training
includes tangent-wise feature contrastive training and prediction consistency
training. That is, the former formulates the features with the same projection
locations as positive examples and vice versa, for the models' awareness of
distortion, while the latter ensures the consistency of cross-model predictions
between the ERP and TP. Moreover, adversarial intra-projection training is
proposed to reduce the inherent gap, between the features of the pinhole images
and those of the ERP and TP images, respectively. Importantly, the TP path can
be freely removed after training, leading to no additional inference cost.
Extensive experiments on two benchmarks show that our DPPASS achieves +1.06$\%$
mIoU increment than the state-of-the-art approaches.
| no_new_dataset | 0.951639 |
2303.14361 | Shao-Yuan Lo | Shao-Yuan Lo, Poojan Oza, Sumanth Chennupati, Alejandro Galindo,
Vishal M. Patel | Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free
Domain Adaptation for Video Semantic Segmentation | Accepted at IEEE/CVF Conference on Computer Vision and Pattern
Recognition (CVPR) 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Unsupervised Domain Adaptation (UDA) of semantic segmentation transfers
labeled source knowledge to an unlabeled target domain by relying on accessing
both the source and target data. However, the access to source data is often
restricted or infeasible in real-world scenarios. Under the source data
restrictive circumstances, UDA is less practical. To address this, recent works
have explored solutions under the Source-Free Domain Adaptation (SFDA) setup,
which aims to adapt a source-trained model to the target domain without
accessing source data. Still, existing SFDA approaches use only image-level
information for adaptation, making them sub-optimal in video applications. This
paper studies SFDA for Video Semantic Segmentation (VSS), where temporal
information is leveraged to address video adaptation. Specifically, we propose
Spatio-Temporal Pixel-Level (STPL) contrastive learning, a novel method that
takes full advantage of spatio-temporal information to tackle the absence of
source data better. STPL explicitly learns semantic correlations among pixels
in the spatio-temporal space, providing strong self-supervision for adaptation
to the unlabeled target domain. Extensive experiments show that STPL achieves
state-of-the-art performance on VSS benchmarks compared to current UDA and SFDA
approaches. Code is available at: https://github.com/shaoyuanlo/STPL
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 05:06:23 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Lo",
"Shao-Yuan",
""
],
[
"Oza",
"Poojan",
""
],
[
"Chennupati",
"Sumanth",
""
],
[
"Galindo",
"Alejandro",
""
],
[
"Patel",
"Vishal M.",
""
]
] | TITLE: Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free
Domain Adaptation for Video Semantic Segmentation
ABSTRACT: Unsupervised Domain Adaptation (UDA) of semantic segmentation transfers
labeled source knowledge to an unlabeled target domain by relying on accessing
both the source and target data. However, the access to source data is often
restricted or infeasible in real-world scenarios. Under the source data
restrictive circumstances, UDA is less practical. To address this, recent works
have explored solutions under the Source-Free Domain Adaptation (SFDA) setup,
which aims to adapt a source-trained model to the target domain without
accessing source data. Still, existing SFDA approaches use only image-level
information for adaptation, making them sub-optimal in video applications. This
paper studies SFDA for Video Semantic Segmentation (VSS), where temporal
information is leveraged to address video adaptation. Specifically, we propose
Spatio-Temporal Pixel-Level (STPL) contrastive learning, a novel method that
takes full advantage of spatio-temporal information to tackle the absence of
source data better. STPL explicitly learns semantic correlations among pixels
in the spatio-temporal space, providing strong self-supervision for adaptation
to the unlabeled target domain. Extensive experiments show that STPL achieves
state-of-the-art performance on VSS benchmarks compared to current UDA and SFDA
approaches. Code is available at: https://github.com/shaoyuanlo/STPL
| no_new_dataset | 0.95253 |
2303.14366 | Akira Kinjo | Akira R. Kinjo and Daphne Teck Ching Lai | Hybrid Fuzzy-Crisp Clustering Algorithm: Theory and Experiments | 41 pages, 13 figures, 10 tables | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | With the membership function being strictly positive, the conventional fuzzy
c-means clustering method sometimes causes imbalanced influence when clusters
of vastly different sizes exist. That is, an outstandingly large cluster drags
to its center all the other clusters, however far they are separated. To solve
this problem, we propose a hybrid fuzzy-crisp clustering algorithm based on a
target function combining linear and quadratic terms of the membership
function. In this algorithm, the membership of a data point to a cluster is
automatically set to exactly zero if the data point is ``sufficiently'' far
from the cluster center. In this paper, we present a new algorithm for hybrid
fuzzy-crisp clustering along with its geometric interpretation. The algorithm
is tested on twenty simulated data generated and five real-world datasets from
the UCI repository and compared with conventional fuzzy and crisp clustering
methods. The proposed algorithm is demonstrated to outperform the conventional
methods on imbalanced datasets and can be competitive on more balanced
datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 05:27:26 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Kinjo",
"Akira R.",
""
],
[
"Lai",
"Daphne Teck Ching",
""
]
] | TITLE: Hybrid Fuzzy-Crisp Clustering Algorithm: Theory and Experiments
ABSTRACT: With the membership function being strictly positive, the conventional fuzzy
c-means clustering method sometimes causes imbalanced influence when clusters
of vastly different sizes exist. That is, an outstandingly large cluster drags
to its center all the other clusters, however far they are separated. To solve
this problem, we propose a hybrid fuzzy-crisp clustering algorithm based on a
target function combining linear and quadratic terms of the membership
function. In this algorithm, the membership of a data point to a cluster is
automatically set to exactly zero if the data point is ``sufficiently'' far
from the cluster center. In this paper, we present a new algorithm for hybrid
fuzzy-crisp clustering along with its geometric interpretation. The algorithm
is tested on twenty simulated data generated and five real-world datasets from
the UCI repository and compared with conventional fuzzy and crisp clustering
methods. The proposed algorithm is demonstrated to outperform the conventional
methods on imbalanced datasets and can be competitive on more balanced
datasets.
| no_new_dataset | 0.951997 |
2303.14368 | Vinoj Jayasundara | Vinoj Jayasundara, Amit Agrawal, Nicolas Heron, Abhinav Shrivastava,
Larry S. Davis | FlexNeRF: Photorealistic Free-viewpoint Rendering of Moving Humans from
Sparse Views | CVPR 2023 | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present FlexNeRF, a method for photorealistic freeviewpoint rendering of
humans in motion from monocular videos. Our approach works well with sparse
views, which is a challenging scenario when the subject is exhibiting
fast/complex motions. We propose a novel approach which jointly optimizes a
canonical time and pose configuration, with a pose-dependent motion field and
pose-independent temporal deformations complementing each other. Thanks to our
novel temporal and cyclic consistency constraints along with additional losses
on intermediate representation such as segmentation, our approach provides high
quality outputs as the observed views become sparser. We empirically
demonstrate that our method significantly outperforms the state-of-the-art on
public benchmark datasets as well as a self-captured fashion dataset. The
project page is available at: https://flex-nerf.github.io/
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 05:47:08 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jayasundara",
"Vinoj",
""
],
[
"Agrawal",
"Amit",
""
],
[
"Heron",
"Nicolas",
""
],
[
"Shrivastava",
"Abhinav",
""
],
[
"Davis",
"Larry S.",
""
]
] | TITLE: FlexNeRF: Photorealistic Free-viewpoint Rendering of Moving Humans from
Sparse Views
ABSTRACT: We present FlexNeRF, a method for photorealistic freeviewpoint rendering of
humans in motion from monocular videos. Our approach works well with sparse
views, which is a challenging scenario when the subject is exhibiting
fast/complex motions. We propose a novel approach which jointly optimizes a
canonical time and pose configuration, with a pose-dependent motion field and
pose-independent temporal deformations complementing each other. Thanks to our
novel temporal and cyclic consistency constraints along with additional losses
on intermediate representation such as segmentation, our approach provides high
quality outputs as the observed views become sparser. We empirically
demonstrate that our method significantly outperforms the state-of-the-art on
public benchmark datasets as well as a self-captured fashion dataset. The
project page is available at: https://flex-nerf.github.io/
| new_dataset | 0.956836 |
2303.14369 | Peng Jin | Peng Jin, Jinfa Huang, Pengfei Xiong, Shangxuan Tian, Chang Liu,
Xiangyang Ji, Li Yuan, Jie Chen | Video-Text as Game Players: Hierarchical Banzhaf Interaction for
Cross-Modal Representation Learning | CVPR 2023 Highlight | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Contrastive learning-based video-language representation learning approaches,
e.g., CLIP, have achieved outstanding performance, which pursue semantic
interaction upon pre-defined video-text pairs. To clarify this coarse-grained
global interaction and move a step further, we have to encounter challenging
shell-breaking interactions for fine-grained cross-modal learning. In this
paper, we creatively model video-text as game players with multivariate
cooperative game theory to wisely handle the uncertainty during fine-grained
semantic interaction with diverse granularity, flexible combination, and vague
intensity. Concretely, we propose Hierarchical Banzhaf Interaction (HBI) to
value possible correspondence between video frames and text words for sensitive
and explainable cross-modal contrast. To efficiently realize the cooperative
game of multiple video frames and multiple text words, the proposed method
clusters the original video frames (text words) and computes the Banzhaf
Interaction between the merged tokens. By stacking token merge modules, we
achieve cooperative games at different semantic levels. Extensive experiments
on commonly used text-video retrieval and video-question answering benchmarks
with superior performances justify the efficacy of our HBI. More encouragingly,
it can also serve as a visualization tool to promote the understanding of
cross-modal interaction, which have a far-reaching impact on the community.
Project page is available at https://jpthu17.github.io/HBI/.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 05:47:52 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jin",
"Peng",
""
],
[
"Huang",
"Jinfa",
""
],
[
"Xiong",
"Pengfei",
""
],
[
"Tian",
"Shangxuan",
""
],
[
"Liu",
"Chang",
""
],
[
"Ji",
"Xiangyang",
""
],
[
"Yuan",
"Li",
""
],
[
"Chen",
"Jie",
""
]
] | TITLE: Video-Text as Game Players: Hierarchical Banzhaf Interaction for
Cross-Modal Representation Learning
ABSTRACT: Contrastive learning-based video-language representation learning approaches,
e.g., CLIP, have achieved outstanding performance, which pursue semantic
interaction upon pre-defined video-text pairs. To clarify this coarse-grained
global interaction and move a step further, we have to encounter challenging
shell-breaking interactions for fine-grained cross-modal learning. In this
paper, we creatively model video-text as game players with multivariate
cooperative game theory to wisely handle the uncertainty during fine-grained
semantic interaction with diverse granularity, flexible combination, and vague
intensity. Concretely, we propose Hierarchical Banzhaf Interaction (HBI) to
value possible correspondence between video frames and text words for sensitive
and explainable cross-modal contrast. To efficiently realize the cooperative
game of multiple video frames and multiple text words, the proposed method
clusters the original video frames (text words) and computes the Banzhaf
Interaction between the merged tokens. By stacking token merge modules, we
achieve cooperative games at different semantic levels. Extensive experiments
on commonly used text-video retrieval and video-question answering benchmarks
with superior performances justify the efficacy of our HBI. More encouragingly,
it can also serve as a visualization tool to promote the understanding of
cross-modal interaction, which have a far-reaching impact on the community.
Project page is available at https://jpthu17.github.io/HBI/.
| no_new_dataset | 0.949059 |
2303.14373 | Jiang Hao | Hao Jiang and Rushan Zhang and Yanning Zhou and Yumeng Wang and Hao
Chen | DoNet: Deep De-overlapping Network for Cytology Instance Segmentation | Accepted by CVPR2023 | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cell instance segmentation in cytology images has significant importance for
biology analysis and cancer screening, while remains challenging due to 1) the
extensive overlapping translucent cell clusters that cause the ambiguous
boundaries, and 2) the confusion of mimics and debris as nuclei. In this work,
we proposed a De-overlapping Network (DoNet) in a decompose-and-recombined
strategy. A Dual-path Region Segmentation Module (DRM) explicitly decomposes
the cell clusters into intersection and complement regions, followed by a
Semantic Consistency-guided Recombination Module (CRM) for integration. To
further introduce the containment relationship of the nucleus in the cytoplasm,
we design a Mask-guided Region Proposal Strategy (MRP) that integrates the cell
attention maps for inner-cell instance prediction. We validate the proposed
approach on ISBI2014 and CPS datasets. Experiments show that our proposed DoNet
significantly outperforms other state-of-the-art (SOTA) cell instance
segmentation methods. The code is available at
https://github.com/DeepDoNet/DoNet.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 06:26:50 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jiang",
"Hao",
""
],
[
"Zhang",
"Rushan",
""
],
[
"Zhou",
"Yanning",
""
],
[
"Wang",
"Yumeng",
""
],
[
"Chen",
"Hao",
""
]
] | TITLE: DoNet: Deep De-overlapping Network for Cytology Instance Segmentation
ABSTRACT: Cell instance segmentation in cytology images has significant importance for
biology analysis and cancer screening, while remains challenging due to 1) the
extensive overlapping translucent cell clusters that cause the ambiguous
boundaries, and 2) the confusion of mimics and debris as nuclei. In this work,
we proposed a De-overlapping Network (DoNet) in a decompose-and-recombined
strategy. A Dual-path Region Segmentation Module (DRM) explicitly decomposes
the cell clusters into intersection and complement regions, followed by a
Semantic Consistency-guided Recombination Module (CRM) for integration. To
further introduce the containment relationship of the nucleus in the cytoplasm,
we design a Mask-guided Region Proposal Strategy (MRP) that integrates the cell
attention maps for inner-cell instance prediction. We validate the proposed
approach on ISBI2014 and CPS datasets. Experiments show that our proposed DoNet
significantly outperforms other state-of-the-art (SOTA) cell instance
segmentation methods. The code is available at
https://github.com/DeepDoNet/DoNet.
| no_new_dataset | 0.949295 |
2303.14375 | Rui Zhang Zhang | Rui Zhang, Yajing Sun, Jingyuan Yang, Wei Peng | Knowledge-augmented Frame Semantic Parsing with Hybrid Prompt-tuning | accepted by ICASSP 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Frame semantics-based approaches have been widely used in semantic parsing
tasks and have become mainstream. It remains challenging to disambiguate frame
representations evoked by target lexical units under different contexts.
Pre-trained Language Models (PLMs) have been used in semantic parsing and
significantly improve the accuracy of neural parsers. However, the PLMs-based
approaches tend to favor collocated patterns presented in the training data,
leading to inaccurate outcomes. The intuition here is to design a mechanism to
optimally use knowledge captured in semantic frames in conjunction with PLMs to
disambiguate frames. We propose a novel Knowledge-Augmented Frame Semantic
Parsing Architecture (KAF-SPA) to enhance semantic representation by
incorporating accurate frame knowledge into PLMs during frame semantic parsing.
Specifically, a Memory-based Knowledge Extraction Module (MKEM) is devised to
select accurate frame knowledge and construct the continuous templates in the
high dimensional vector space. Moreover, we design a Task-oriented Knowledge
Probing Module (TKPM) using hybrid prompts (in terms of continuous and discrete
prompts) to incorporate the selected knowledge into the PLMs and adapt PLMs to
the tasks of frame and argument identification. Experimental results on two
public FrameNet datasets demonstrate that our method significantly outperforms
strong baselines (by more than +3$\%$ in F1), achieving state-of-art results on
the current benchmark. Ablation studies verify the effectiveness of KAF-SPA.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 06:41:19 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Rui",
""
],
[
"Sun",
"Yajing",
""
],
[
"Yang",
"Jingyuan",
""
],
[
"Peng",
"Wei",
""
]
] | TITLE: Knowledge-augmented Frame Semantic Parsing with Hybrid Prompt-tuning
ABSTRACT: Frame semantics-based approaches have been widely used in semantic parsing
tasks and have become mainstream. It remains challenging to disambiguate frame
representations evoked by target lexical units under different contexts.
Pre-trained Language Models (PLMs) have been used in semantic parsing and
significantly improve the accuracy of neural parsers. However, the PLMs-based
approaches tend to favor collocated patterns presented in the training data,
leading to inaccurate outcomes. The intuition here is to design a mechanism to
optimally use knowledge captured in semantic frames in conjunction with PLMs to
disambiguate frames. We propose a novel Knowledge-Augmented Frame Semantic
Parsing Architecture (KAF-SPA) to enhance semantic representation by
incorporating accurate frame knowledge into PLMs during frame semantic parsing.
Specifically, a Memory-based Knowledge Extraction Module (MKEM) is devised to
select accurate frame knowledge and construct the continuous templates in the
high dimensional vector space. Moreover, we design a Task-oriented Knowledge
Probing Module (TKPM) using hybrid prompts (in terms of continuous and discrete
prompts) to incorporate the selected knowledge into the PLMs and adapt PLMs to
the tasks of frame and argument identification. Experimental results on two
public FrameNet datasets demonstrate that our method significantly outperforms
strong baselines (by more than +3$\%$ in F1), achieving state-of-art results on
the current benchmark. Ablation studies verify the effectiveness of KAF-SPA.
| no_new_dataset | 0.94868 |
2303.14376 | Hongyu Sun | Hongyu Sun, Yongcai Wang, Xudong Cai, Xuewei Bai and Deying Li | ViPFormer: Efficient Vision-and-Pointcloud Transformer for Unsupervised
Pointcloud Understanding | 9 pages, 4 figures, 7 tables; accepted by ICRA 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recently, a growing number of work design unsupervised paradigms for point
cloud processing to alleviate the limitation of expensive manual annotation and
poor transferability of supervised methods. Among them, CrossPoint follows the
contrastive learning framework and exploits image and point cloud data for
unsupervised point cloud understanding. Although the promising performance is
presented, the unbalanced architecture makes it unnecessarily complex and
inefficient. For example, the image branch in CrossPoint is $\sim$8.3x heavier
than the point cloud branch leading to higher complexity and latency. To
address this problem, in this paper, we propose a lightweight
Vision-and-Pointcloud Transformer (ViPFormer) to unify image and point cloud
processing in a single architecture. ViPFormer learns in an unsupervised manner
by optimizing intra-modal and cross-modal contrastive objectives. Then the
pretrained model is transferred to various downstream tasks, including 3D shape
classification and semantic segmentation. Experiments on different datasets
show ViPFormer surpasses previous state-of-the-art unsupervised methods with
higher accuracy, lower model complexity and runtime latency. Finally, the
effectiveness of each component in ViPFormer is validated by extensive ablation
studies. The implementation of the proposed method is available at
https://github.com/auniquesun/ViPFormer.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 06:47:12 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Sun",
"Hongyu",
""
],
[
"Wang",
"Yongcai",
""
],
[
"Cai",
"Xudong",
""
],
[
"Bai",
"Xuewei",
""
],
[
"Li",
"Deying",
""
]
] | TITLE: ViPFormer: Efficient Vision-and-Pointcloud Transformer for Unsupervised
Pointcloud Understanding
ABSTRACT: Recently, a growing number of work design unsupervised paradigms for point
cloud processing to alleviate the limitation of expensive manual annotation and
poor transferability of supervised methods. Among them, CrossPoint follows the
contrastive learning framework and exploits image and point cloud data for
unsupervised point cloud understanding. Although the promising performance is
presented, the unbalanced architecture makes it unnecessarily complex and
inefficient. For example, the image branch in CrossPoint is $\sim$8.3x heavier
than the point cloud branch leading to higher complexity and latency. To
address this problem, in this paper, we propose a lightweight
Vision-and-Pointcloud Transformer (ViPFormer) to unify image and point cloud
processing in a single architecture. ViPFormer learns in an unsupervised manner
by optimizing intra-modal and cross-modal contrastive objectives. Then the
pretrained model is transferred to various downstream tasks, including 3D shape
classification and semantic segmentation. Experiments on different datasets
show ViPFormer surpasses previous state-of-the-art unsupervised methods with
higher accuracy, lower model complexity and runtime latency. Finally, the
effectiveness of each component in ViPFormer is validated by extensive ablation
studies. The implementation of the proposed method is available at
https://github.com/auniquesun/ViPFormer.
| no_new_dataset | 0.949623 |
2303.14377 | Chenchen Xu | Chenchen Xu and Min Zhou and Tiezheng Ge and Yuning Jiang and Weiwei
Xu | Unsupervised Domain Adaption with Pixel-level Discriminator for
Image-aware Layout Generation | 8 pages, 4 figures, 7 tables, accepted by CVPR2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Layout is essential for graphic design and poster generation. Recently,
applying deep learning models to generate layouts has attracted increasing
attention. This paper focuses on using the GAN-based model conditioned on image
contents to generate advertising poster graphic layouts, which requires an
advertising poster layout dataset with paired product images and graphic
layouts. However, the paired images and layouts in the existing dataset are
collected by inpainting and annotating posters, respectively. There exists a
domain gap between inpainted posters (source domain data) and clean product
images (target domain data). Therefore, this paper combines unsupervised domain
adaption techniques to design a GAN with a novel pixel-level discriminator
(PD), called PDA-GAN, to generate graphic layouts according to image contents.
The PD is connected to the shallow level feature map and computes the GAN loss
for each input-image pixel. Both quantitative and qualitative evaluations
demonstrate that PDA-GAN can achieve state-of-the-art performances and generate
high-quality image-aware graphic layouts for advertising posters.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 06:50:22 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Xu",
"Chenchen",
""
],
[
"Zhou",
"Min",
""
],
[
"Ge",
"Tiezheng",
""
],
[
"Jiang",
"Yuning",
""
],
[
"Xu",
"Weiwei",
""
]
] | TITLE: Unsupervised Domain Adaption with Pixel-level Discriminator for
Image-aware Layout Generation
ABSTRACT: Layout is essential for graphic design and poster generation. Recently,
applying deep learning models to generate layouts has attracted increasing
attention. This paper focuses on using the GAN-based model conditioned on image
contents to generate advertising poster graphic layouts, which requires an
advertising poster layout dataset with paired product images and graphic
layouts. However, the paired images and layouts in the existing dataset are
collected by inpainting and annotating posters, respectively. There exists a
domain gap between inpainted posters (source domain data) and clean product
images (target domain data). Therefore, this paper combines unsupervised domain
adaption techniques to design a GAN with a novel pixel-level discriminator
(PD), called PDA-GAN, to generate graphic layouts according to image contents.
The PD is connected to the shallow level feature map and computes the GAN loss
for each input-image pixel. Both quantitative and qualitative evaluations
demonstrate that PDA-GAN can achieve state-of-the-art performances and generate
high-quality image-aware graphic layouts for advertising posters.
| no_new_dataset | 0.947381 |
2303.14378 | Soonmin Hwang | Kwonyoung Ryu, Soonmin Hwang, Jaesik Park | Instant Domain Augmentation for LiDAR Semantic Segmentation | Accepted to CVPR 2023. Project page:
http://cvlab.postech.ac.kr/research/LiDomAug | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite the increasing popularity of LiDAR sensors, perception algorithms
using 3D LiDAR data struggle with the 'sensor-bias problem'. Specifically, the
performance of perception algorithms significantly drops when an unseen
specification of LiDAR sensor is applied at test time due to the domain
discrepancy. This paper presents a fast and flexible LiDAR augmentation method
for the semantic segmentation task, called 'LiDomAug'. It aggregates raw LiDAR
scans and creates a LiDAR scan of any configurations with the consideration of
dynamic distortion and occlusion, resulting in instant domain augmentation. Our
on-demand augmentation module runs at 330 FPS, so it can be seamlessly
integrated into the data loader in the learning framework. In our experiments,
learning-based approaches aided with the proposed LiDomAug are less affected by
the sensor-bias issue and achieve new state-of-the-art domain adaptation
performances on SemanticKITTI and nuScenes dataset without the use of the
target domain data. We also present a sensor-agnostic model that faithfully
works on the various LiDAR configurations.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 06:59:12 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ryu",
"Kwonyoung",
""
],
[
"Hwang",
"Soonmin",
""
],
[
"Park",
"Jaesik",
""
]
] | TITLE: Instant Domain Augmentation for LiDAR Semantic Segmentation
ABSTRACT: Despite the increasing popularity of LiDAR sensors, perception algorithms
using 3D LiDAR data struggle with the 'sensor-bias problem'. Specifically, the
performance of perception algorithms significantly drops when an unseen
specification of LiDAR sensor is applied at test time due to the domain
discrepancy. This paper presents a fast and flexible LiDAR augmentation method
for the semantic segmentation task, called 'LiDomAug'. It aggregates raw LiDAR
scans and creates a LiDAR scan of any configurations with the consideration of
dynamic distortion and occlusion, resulting in instant domain augmentation. Our
on-demand augmentation module runs at 330 FPS, so it can be seamlessly
integrated into the data loader in the learning framework. In our experiments,
learning-based approaches aided with the proposed LiDomAug are less affected by
the sensor-bias issue and achieve new state-of-the-art domain adaptation
performances on SemanticKITTI and nuScenes dataset without the use of the
target domain data. We also present a sensor-agnostic model that faithfully
works on the various LiDAR configurations.
| no_new_dataset | 0.950549 |
2303.14381 | Thomas Le | Phuong D. Nguyen, Thinh D. Le, Duong Q. Nguyen, Thanh Q. Nguyen,
Li-Wei Chou, H. Nguyen-Xuan | 3D Facial Imperfection Regeneration: Deep learning approach and 3D
printing prototypes | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This study explores the potential of a fully convolutional mesh autoencoder
model for regenerating 3D nature faces with the presence of imperfect areas. We
utilize deep learning approaches in graph processing and analysis to
investigate the capabilities model in recreating a filling part for facial
scars. Our approach in dataset creation is able to generate a facial scar
rationally in a virtual space that corresponds to the unique circumstances.
Especially, we propose a new method which is named 3D Facial Imperfection
Regeneration(3D-FaIR) for reproducing a complete face reconstruction based on
the remaining features of the patient face. To further enhance the applicable
capacity of the present research, we develop an improved outlier technique to
separate the wounds of patients and provide appropriate wound cover models.
Also, a Cir3D-FaIR dataset of imperfect faces and open codes was released at
https://github.com/SIMOGroup/3DFaIR. Our findings demonstrate the potential of
the proposed approach to help patients recover more quickly and safely through
convenient techniques. We hope that this research can contribute to the
development of new products and innovative solutions for facial scar
regeneration.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 07:12:33 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Nguyen",
"Phuong D.",
""
],
[
"Le",
"Thinh D.",
""
],
[
"Nguyen",
"Duong Q.",
""
],
[
"Nguyen",
"Thanh Q.",
""
],
[
"Chou",
"Li-Wei",
""
],
[
"Nguyen-Xuan",
"H.",
""
]
] | TITLE: 3D Facial Imperfection Regeneration: Deep learning approach and 3D
printing prototypes
ABSTRACT: This study explores the potential of a fully convolutional mesh autoencoder
model for regenerating 3D nature faces with the presence of imperfect areas. We
utilize deep learning approaches in graph processing and analysis to
investigate the capabilities model in recreating a filling part for facial
scars. Our approach in dataset creation is able to generate a facial scar
rationally in a virtual space that corresponds to the unique circumstances.
Especially, we propose a new method which is named 3D Facial Imperfection
Regeneration(3D-FaIR) for reproducing a complete face reconstruction based on
the remaining features of the patient face. To further enhance the applicable
capacity of the present research, we develop an improved outlier technique to
separate the wounds of patients and provide appropriate wound cover models.
Also, a Cir3D-FaIR dataset of imperfect faces and open codes was released at
https://github.com/SIMOGroup/3DFaIR. Our findings demonstrate the potential of
the proposed approach to help patients recover more quickly and safely through
convenient techniques. We hope that this research can contribute to the
development of new products and innovative solutions for facial scar
regeneration.
| new_dataset | 0.957517 |
2303.14384 | Zikun Zhou | Zikun Zhou, Kaige Mao, Wenjie Pei, Hongpeng Wang, Yaowei Wang, Zhenyu
He | Reliability-Hierarchical Memory Network for Scribble-Supervised Video
Object Segmentation | This project is available at
https://github.com/mkg1204/RHMNet-for-SSVOS | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper aims to solve the video object segmentation (VOS) task in a
scribble-supervised manner, in which VOS models are not only trained by the
sparse scribble annotations but also initialized with the sparse target
scribbles for inference. Thus, the annotation burdens for both training and
initialization can be substantially lightened. The difficulties of
scribble-supervised VOS lie in two aspects. On the one hand, it requires the
powerful ability to learn from the sparse scribble annotations during training.
On the other hand, it demands strong reasoning capability during inference
given only a sparse initial target scribble. In this work, we propose a
Reliability-Hierarchical Memory Network (RHMNet) to predict the target mask in
a step-wise expanding strategy w.r.t. the memory reliability level. To be
specific, RHMNet first only uses the memory in the high-reliability level to
locate the region with high reliability belonging to the target, which is
highly similar to the initial target scribble. Then it expands the located
high-reliability region to the entire target conditioned on the region itself
and the memories in all reliability levels. Besides, we propose a
scribble-supervised learning mechanism to facilitate the learning of our model
to predict dense results. It mines the pixel-level relation within the single
frame and the frame-level relation within the sequence to take full advantage
of the scribble annotations in sequence training samples. The favorable
performance on two popular benchmarks demonstrates that our method is
promising.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 07:21:40 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhou",
"Zikun",
""
],
[
"Mao",
"Kaige",
""
],
[
"Pei",
"Wenjie",
""
],
[
"Wang",
"Hongpeng",
""
],
[
"Wang",
"Yaowei",
""
],
[
"He",
"Zhenyu",
""
]
] | TITLE: Reliability-Hierarchical Memory Network for Scribble-Supervised Video
Object Segmentation
ABSTRACT: This paper aims to solve the video object segmentation (VOS) task in a
scribble-supervised manner, in which VOS models are not only trained by the
sparse scribble annotations but also initialized with the sparse target
scribbles for inference. Thus, the annotation burdens for both training and
initialization can be substantially lightened. The difficulties of
scribble-supervised VOS lie in two aspects. On the one hand, it requires the
powerful ability to learn from the sparse scribble annotations during training.
On the other hand, it demands strong reasoning capability during inference
given only a sparse initial target scribble. In this work, we propose a
Reliability-Hierarchical Memory Network (RHMNet) to predict the target mask in
a step-wise expanding strategy w.r.t. the memory reliability level. To be
specific, RHMNet first only uses the memory in the high-reliability level to
locate the region with high reliability belonging to the target, which is
highly similar to the initial target scribble. Then it expands the located
high-reliability region to the entire target conditioned on the region itself
and the memories in all reliability levels. Besides, we propose a
scribble-supervised learning mechanism to facilitate the learning of our model
to predict dense results. It mines the pixel-level relation within the single
frame and the frame-level relation within the sequence to take full advantage
of the scribble annotations in sequence training samples. The favorable
performance on two popular benchmarks demonstrates that our method is
promising.
| no_new_dataset | 0.954984 |
2303.14386 | Hwanjun Song | Hwanjun Song, Jihwan Bang | Prompt-Guided Transformers for End-to-End Open-Vocabulary Object
Detection | version 1 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Prompt-OVD is an efficient and effective framework for open-vocabulary object
detection that utilizes class embeddings from CLIP as prompts, guiding the
Transformer decoder to detect objects in both base and novel classes.
Additionally, our novel RoI-based masked attention and RoI pruning techniques
help leverage the zero-shot classification ability of the Vision
Transformer-based CLIP, resulting in improved detection performance at minimal
computational cost. Our experiments on the OV-COCO and OVLVIS datasets
demonstrate that Prompt-OVD achieves an impressive 21.2 times faster inference
speed than the first end-to-end open-vocabulary detection method (OV-DETR),
while also achieving higher APs than four two-stage-based methods operating
within similar inference time ranges. Code will be made available soon.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 07:31:08 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Song",
"Hwanjun",
""
],
[
"Bang",
"Jihwan",
""
]
] | TITLE: Prompt-Guided Transformers for End-to-End Open-Vocabulary Object
Detection
ABSTRACT: Prompt-OVD is an efficient and effective framework for open-vocabulary object
detection that utilizes class embeddings from CLIP as prompts, guiding the
Transformer decoder to detect objects in both base and novel classes.
Additionally, our novel RoI-based masked attention and RoI pruning techniques
help leverage the zero-shot classification ability of the Vision
Transformer-based CLIP, resulting in improved detection performance at minimal
computational cost. Our experiments on the OV-COCO and OVLVIS datasets
demonstrate that Prompt-OVD achieves an impressive 21.2 times faster inference
speed than the first end-to-end open-vocabulary detection method (OV-DETR),
while also achieving higher APs than four two-stage-based methods operating
within similar inference time ranges. Code will be made available soon.
| no_new_dataset | 0.948251 |
2303.14402 | Yorick Broens | Anantha Sai Hariharan Vinjarapu and Yorick Broens and Hans Butler and
Roland T\'oth | Exploring the use of deep learning in task-flexible ILC | 6 pages, 5 figures, Regular paper ACC conference 2023 | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | Growing demands in today's industry results in increasingly stringent
performance and throughput specifications. For accurate positioning of
high-precision motion systems, feedforward control plays a crucial role.
Nonetheless, conventional model-based feedforward approaches are no longer
sufficient to satisfy the challenging performance requirements. An attractive
method for systems with repetitive motion tasks is iterative learning control
(ILC) due to its superior performance. However, for systems with non-repetitive
motion tasks, ILC is {generally} not applicable, {despite of some recent
promising advances}. In this paper, we aim to explore the use of deep learning
to address the task flexibility constraint of ILC. For this purpose, a novel
Task Analogy based Imitation Learning (TAIL)-ILC approach is developed. To
benchmark the performance of the proposed approach, a simulation study is
presented which compares the TAIL-ILC to classical model-based feedforward
strategies and existing learning-based approaches, such as neural network based
feedforward learning.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 08:41:38 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Vinjarapu",
"Anantha Sai Hariharan",
""
],
[
"Broens",
"Yorick",
""
],
[
"Butler",
"Hans",
""
],
[
"Tóth",
"Roland",
""
]
] | TITLE: Exploring the use of deep learning in task-flexible ILC
ABSTRACT: Growing demands in today's industry results in increasingly stringent
performance and throughput specifications. For accurate positioning of
high-precision motion systems, feedforward control plays a crucial role.
Nonetheless, conventional model-based feedforward approaches are no longer
sufficient to satisfy the challenging performance requirements. An attractive
method for systems with repetitive motion tasks is iterative learning control
(ILC) due to its superior performance. However, for systems with non-repetitive
motion tasks, ILC is {generally} not applicable, {despite of some recent
promising advances}. In this paper, we aim to explore the use of deep learning
to address the task flexibility constraint of ILC. For this purpose, a novel
Task Analogy based Imitation Learning (TAIL)-ILC approach is developed. To
benchmark the performance of the proposed approach, a simulation study is
presented which compares the TAIL-ILC to classical model-based feedforward
strategies and existing learning-based approaches, such as neural network based
feedforward learning.
| no_new_dataset | 0.940463 |
2303.14404 | Muhammad Akhtar Munir | Muhammad Akhtar Munir and Muhammad Haris Khan and Salman Khan and
Fahad Shahbaz Khan | Bridging Precision and Confidence: A Train-Time Loss for Calibrating
Object Detection | Accepted at CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deep neural networks (DNNs) have enabled astounding progress in several
vision-based problems. Despite showing high predictive accuracy, recently,
several works have revealed that they tend to provide overconfident predictions
and thus are poorly calibrated. The majority of the works addressing the
miscalibration of DNNs fall under the scope of classification and consider only
in-domain predictions. However, there is little to no progress in studying the
calibration of DNN-based object detection models, which are central to many
vision-based safety-critical applications. In this paper, inspired by the
train-time calibration methods, we propose a novel auxiliary loss formulation
that explicitly aims to align the class confidence of bounding boxes with the
accurateness of predictions (i.e. precision). Since the original formulation of
our loss depends on the counts of true positives and false positives in a
minibatch, we develop a differentiable proxy of our loss that can be used
during training with other application-specific loss functions. We perform
extensive experiments on challenging in-domain and out-domain scenarios with
six benchmark datasets including MS-COCO, Cityscapes, Sim10k, and BDD100k. Our
results reveal that our train-time loss surpasses strong calibration baselines
in reducing calibration error for both in and out-domain scenarios. Our source
code and pre-trained models are available at
https://github.com/akhtarvision/bpc_calibration
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 08:56:21 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Munir",
"Muhammad Akhtar",
""
],
[
"Khan",
"Muhammad Haris",
""
],
[
"Khan",
"Salman",
""
],
[
"Khan",
"Fahad Shahbaz",
""
]
] | TITLE: Bridging Precision and Confidence: A Train-Time Loss for Calibrating
Object Detection
ABSTRACT: Deep neural networks (DNNs) have enabled astounding progress in several
vision-based problems. Despite showing high predictive accuracy, recently,
several works have revealed that they tend to provide overconfident predictions
and thus are poorly calibrated. The majority of the works addressing the
miscalibration of DNNs fall under the scope of classification and consider only
in-domain predictions. However, there is little to no progress in studying the
calibration of DNN-based object detection models, which are central to many
vision-based safety-critical applications. In this paper, inspired by the
train-time calibration methods, we propose a novel auxiliary loss formulation
that explicitly aims to align the class confidence of bounding boxes with the
accurateness of predictions (i.e. precision). Since the original formulation of
our loss depends on the counts of true positives and false positives in a
minibatch, we develop a differentiable proxy of our loss that can be used
during training with other application-specific loss functions. We perform
extensive experiments on challenging in-domain and out-domain scenarios with
six benchmark datasets including MS-COCO, Cityscapes, Sim10k, and BDD100k. Our
results reveal that our train-time loss surpasses strong calibration baselines
in reducing calibration error for both in and out-domain scenarios. Our source
code and pre-trained models are available at
https://github.com/akhtarvision/bpc_calibration
| no_new_dataset | 0.940517 |
2303.14407 | Yiqian Wu | Yiqian Wu, Jing Zhang, Hongbo Fu, Xiaogang Jin | LPFF: A Portrait Dataset for Face Generators Across Large Poses | 9 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The creation of 2D realistic facial images and 3D face shapes using
generative networks has been a hot topic in recent years. Existing face
generators exhibit exceptional performance on faces in small to medium poses
(with respect to frontal faces) but struggle to produce realistic results for
large poses. The distorted rendering results on large poses in 3D-aware
generators further show that the generated 3D face shapes are far from the
distribution of 3D faces in reality. We find that the above issues are caused
by the training dataset's pose imbalance.
In this paper, we present LPFF, a large-pose Flickr face dataset comprised of
19,590 high-quality real large-pose portrait images. We utilize our dataset to
train a 2D face generator that can process large-pose face images, as well as a
3D-aware generator that can generate realistic human face geometry. To better
validate our pose-conditional 3D-aware generators, we develop a new FID measure
to evaluate the 3D-level performance. Through this novel FID measure and other
experiments, we show that LPFF can help 2D face generators extend their latent
space and better manipulate the large-pose data, and help 3D-aware face
generators achieve better view consistency and more realistic 3D reconstruction
results.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 09:07:36 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wu",
"Yiqian",
""
],
[
"Zhang",
"Jing",
""
],
[
"Fu",
"Hongbo",
""
],
[
"Jin",
"Xiaogang",
""
]
] | TITLE: LPFF: A Portrait Dataset for Face Generators Across Large Poses
ABSTRACT: The creation of 2D realistic facial images and 3D face shapes using
generative networks has been a hot topic in recent years. Existing face
generators exhibit exceptional performance on faces in small to medium poses
(with respect to frontal faces) but struggle to produce realistic results for
large poses. The distorted rendering results on large poses in 3D-aware
generators further show that the generated 3D face shapes are far from the
distribution of 3D faces in reality. We find that the above issues are caused
by the training dataset's pose imbalance.
In this paper, we present LPFF, a large-pose Flickr face dataset comprised of
19,590 high-quality real large-pose portrait images. We utilize our dataset to
train a 2D face generator that can process large-pose face images, as well as a
3D-aware generator that can generate realistic human face geometry. To better
validate our pose-conditional 3D-aware generators, we develop a new FID measure
to evaluate the 3D-level performance. Through this novel FID measure and other
experiments, we show that LPFF can help 2D face generators extend their latent
space and better manipulate the large-pose data, and help 3D-aware face
generators achieve better view consistency and more realistic 3D reconstruction
results.
| new_dataset | 0.959573 |
2303.14408 | Ziqin Wang | Ziqin Wang, Bowen Cheng, Lichen Zhao, Dong Xu, Yang Tang, Lu Sheng | VL-SAT: Visual-Linguistic Semantics Assisted Training for 3D Semantic
Scene Graph Prediction in Point Cloud | CVPR2023 Highlight | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The task of 3D semantic scene graph (3DSSG) prediction in the point cloud is
challenging since (1) the 3D point cloud only captures geometric structures
with limited semantics compared to 2D images, and (2) long-tailed relation
distribution inherently hinders the learning of unbiased prediction. Since 2D
images provide rich semantics and scene graphs are in nature coped with
languages, in this study, we propose Visual-Linguistic Semantics Assisted
Training (VL-SAT) scheme that can significantly empower 3DSSG prediction models
with discrimination about long-tailed and ambiguous semantic relations. The key
idea is to train a powerful multi-modal oracle model to assist the 3D model.
This oracle learns reliable structural representations based on semantics from
vision, language, and 3D geometry, and its benefits can be heterogeneously
passed to the 3D model during the training stage. By effectively utilizing
visual-linguistic semantics in training, our VL-SAT can significantly boost
common 3DSSG prediction models, such as SGFN and SGGpoint, only with 3D inputs
in the inference stage, especially when dealing with tail relation triplets.
Comprehensive evaluations and ablation studies on the 3DSSG dataset have
validated the effectiveness of the proposed scheme. Code is available at
https://github.com/wz7in/CVPR2023-VLSAT.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 09:14:18 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Ziqin",
""
],
[
"Cheng",
"Bowen",
""
],
[
"Zhao",
"Lichen",
""
],
[
"Xu",
"Dong",
""
],
[
"Tang",
"Yang",
""
],
[
"Sheng",
"Lu",
""
]
] | TITLE: VL-SAT: Visual-Linguistic Semantics Assisted Training for 3D Semantic
Scene Graph Prediction in Point Cloud
ABSTRACT: The task of 3D semantic scene graph (3DSSG) prediction in the point cloud is
challenging since (1) the 3D point cloud only captures geometric structures
with limited semantics compared to 2D images, and (2) long-tailed relation
distribution inherently hinders the learning of unbiased prediction. Since 2D
images provide rich semantics and scene graphs are in nature coped with
languages, in this study, we propose Visual-Linguistic Semantics Assisted
Training (VL-SAT) scheme that can significantly empower 3DSSG prediction models
with discrimination about long-tailed and ambiguous semantic relations. The key
idea is to train a powerful multi-modal oracle model to assist the 3D model.
This oracle learns reliable structural representations based on semantics from
vision, language, and 3D geometry, and its benefits can be heterogeneously
passed to the 3D model during the training stage. By effectively utilizing
visual-linguistic semantics in training, our VL-SAT can significantly boost
common 3DSSG prediction models, such as SGFN and SGGpoint, only with 3D inputs
in the inference stage, especially when dealing with tail relation triplets.
Comprehensive evaluations and ablation studies on the 3DSSG dataset have
validated the effectiveness of the proposed scheme. Code is available at
https://github.com/wz7in/CVPR2023-VLSAT.
| no_new_dataset | 0.949529 |
2303.14411 | Tatiana Tommasi | Leonardo Iurada, Silvia Bucci, Timothy M. Hospedales, Tatiana Tommasi | Fairness meets Cross-Domain Learning: a new perspective on Models and
Metrics | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning-based recognition systems are deployed at scale for several
real-world applications that inevitably involve our social life. Although being
of great support when making complex decisions, they might capture spurious
data correlations and leverage sensitive attributes (e.g. age, gender,
ethnicity). How to factor out this information while keeping a high prediction
performance is a task with still several open questions, many of which are
shared with those of the domain adaptation and generalization literature which
focuses on avoiding visual domain biases. In this work, we propose an in-depth
study of the relationship between cross-domain learning (CD) and model fairness
by introducing a benchmark on face and medical images spanning several
demographic groups as well as classification and localization tasks. After
having highlighted the limits of the current evaluation metrics, we introduce a
new Harmonic Fairness (HF) score to assess jointly how fair and accurate every
model is with respect to a reference baseline. Our study covers 14 CD
approaches alongside three state-of-the-art fairness algorithms and shows how
the former can outperform the latter. Overall, our work paves the way for a
more systematic analysis of fairness problems in computer vision. Code
available at: https://github.com/iurada/fairness_crossdomain
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 09:34:05 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Iurada",
"Leonardo",
""
],
[
"Bucci",
"Silvia",
""
],
[
"Hospedales",
"Timothy M.",
""
],
[
"Tommasi",
"Tatiana",
""
]
] | TITLE: Fairness meets Cross-Domain Learning: a new perspective on Models and
Metrics
ABSTRACT: Deep learning-based recognition systems are deployed at scale for several
real-world applications that inevitably involve our social life. Although being
of great support when making complex decisions, they might capture spurious
data correlations and leverage sensitive attributes (e.g. age, gender,
ethnicity). How to factor out this information while keeping a high prediction
performance is a task with still several open questions, many of which are
shared with those of the domain adaptation and generalization literature which
focuses on avoiding visual domain biases. In this work, we propose an in-depth
study of the relationship between cross-domain learning (CD) and model fairness
by introducing a benchmark on face and medical images spanning several
demographic groups as well as classification and localization tasks. After
having highlighted the limits of the current evaluation metrics, we introduce a
new Harmonic Fairness (HF) score to assess jointly how fair and accurate every
model is with respect to a reference baseline. Our study covers 14 CD
approaches alongside three state-of-the-art fairness algorithms and shows how
the former can outperform the latter. Overall, our work paves the way for a
more systematic analysis of fairness problems in computer vision. Code
available at: https://github.com/iurada/fairness_crossdomain
| no_new_dataset | 0.936168 |
2303.14429 | Yaroslav Zharov | Yaroslav Zharov, Evelina Ametova, Rebecca Spiecker, Tilo Baumbach,
Genoveva Burca, Vincent Heuveline | Shot Noise Reduction in Radiographic and Tomographic Multi-Channel
Imaging with Self-Supervised Deep Learning | To be submitted to Optics Express | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Noise is an important issue for radiographic and tomographic imaging
techniques. It becomes particularly critical in applications where additional
constraints force a strong reduction of the Signal-to-Noise Ratio (SNR) per
image. These constraints may result from limitations on the maximum available
flux or permissible dose and the associated restriction on exposure time.
Often, a high SNR per image is traded for the ability to distribute a given
total exposure capacity per pixel over multiple channels, thus obtaining
additional information about the object by the same total exposure time. These
can be energy channels in the case of spectroscopic imaging or time channels in
the case of time-resolved imaging. In this paper, we report on a method for
improving the quality of noisy multi-channel (time or energy-resolved) imaging
datasets. The method relies on the recent Noise2Noise (N2N) self-supervised
denoising approach that learns to predict a noise-free signal without access to
noise-free data. N2N in turn requires drawing pairs of samples from a data
distribution sharing identical signals while being exposed to different samples
of random noise. The method is applicable if adjacent channels share enough
information to provide images with similar enough information but independent
noise. We demonstrate several representative case studies, namely spectroscopic
(k-edge) X-ray tomography, in vivo X-ray cine-radiography, and
energy-dispersive (Bragg edge) neutron tomography. In all cases, the N2N method
shows dramatic improvement and outperforms conventional denoising methods. For
such imaging techniques, the method can therefore significantly improve image
quality, or maintain image quality with further reduced exposure time per
image.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 10:33:41 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zharov",
"Yaroslav",
""
],
[
"Ametova",
"Evelina",
""
],
[
"Spiecker",
"Rebecca",
""
],
[
"Baumbach",
"Tilo",
""
],
[
"Burca",
"Genoveva",
""
],
[
"Heuveline",
"Vincent",
""
]
] | TITLE: Shot Noise Reduction in Radiographic and Tomographic Multi-Channel
Imaging with Self-Supervised Deep Learning
ABSTRACT: Noise is an important issue for radiographic and tomographic imaging
techniques. It becomes particularly critical in applications where additional
constraints force a strong reduction of the Signal-to-Noise Ratio (SNR) per
image. These constraints may result from limitations on the maximum available
flux or permissible dose and the associated restriction on exposure time.
Often, a high SNR per image is traded for the ability to distribute a given
total exposure capacity per pixel over multiple channels, thus obtaining
additional information about the object by the same total exposure time. These
can be energy channels in the case of spectroscopic imaging or time channels in
the case of time-resolved imaging. In this paper, we report on a method for
improving the quality of noisy multi-channel (time or energy-resolved) imaging
datasets. The method relies on the recent Noise2Noise (N2N) self-supervised
denoising approach that learns to predict a noise-free signal without access to
noise-free data. N2N in turn requires drawing pairs of samples from a data
distribution sharing identical signals while being exposed to different samples
of random noise. The method is applicable if adjacent channels share enough
information to provide images with similar enough information but independent
noise. We demonstrate several representative case studies, namely spectroscopic
(k-edge) X-ray tomography, in vivo X-ray cine-radiography, and
energy-dispersive (Bragg edge) neutron tomography. In all cases, the N2N method
shows dramatic improvement and outperforms conventional denoising methods. For
such imaging techniques, the method can therefore significantly improve image
quality, or maintain image quality with further reduced exposure time per
image.
| no_new_dataset | 0.955026 |
2303.14433 | Jihyo Kim | Jihyo Kim, Jeonghyeon Kim, Sangheum Hwang | Deep Active Learning with Contrastive Learning Under Realistic Data Pool
Assumptions | AAAI 2023 Workshop on Practical Deep Learning in the Wild | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Active learning aims to identify the most informative data from an unlabeled
data pool that enables a model to reach the desired accuracy rapidly. This
benefits especially deep neural networks which generally require a huge number
of labeled samples to achieve high performance. Most existing active learning
methods have been evaluated in an ideal setting where only samples relevant to
the target task, i.e., in-distribution samples, exist in an unlabeled data
pool. A data pool gathered from the wild, however, is likely to include samples
that are irrelevant to the target task at all and/or too ambiguous to assign a
single class label even for the oracle. We argue that assuming an unlabeled
data pool consisting of samples from various distributions is more realistic.
In this work, we introduce new active learning benchmarks that include
ambiguous, task-irrelevant out-of-distribution as well as in-distribution
samples. We also propose an active learning method designed to acquire
informative in-distribution samples in priority. The proposed method leverages
both labeled and unlabeled data pools and selects samples from clusters on the
feature space constructed via contrastive learning. Experimental results
demonstrate that the proposed method requires a lower annotation budget than
existing active learning methods to reach the same level of accuracy.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 10:46:10 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Kim",
"Jihyo",
""
],
[
"Kim",
"Jeonghyeon",
""
],
[
"Hwang",
"Sangheum",
""
]
] | TITLE: Deep Active Learning with Contrastive Learning Under Realistic Data Pool
Assumptions
ABSTRACT: Active learning aims to identify the most informative data from an unlabeled
data pool that enables a model to reach the desired accuracy rapidly. This
benefits especially deep neural networks which generally require a huge number
of labeled samples to achieve high performance. Most existing active learning
methods have been evaluated in an ideal setting where only samples relevant to
the target task, i.e., in-distribution samples, exist in an unlabeled data
pool. A data pool gathered from the wild, however, is likely to include samples
that are irrelevant to the target task at all and/or too ambiguous to assign a
single class label even for the oracle. We argue that assuming an unlabeled
data pool consisting of samples from various distributions is more realistic.
In this work, we introduce new active learning benchmarks that include
ambiguous, task-irrelevant out-of-distribution as well as in-distribution
samples. We also propose an active learning method designed to acquire
informative in-distribution samples in priority. The proposed method leverages
both labeled and unlabeled data pools and selects samples from clusters on the
feature space constructed via contrastive learning. Experimental results
demonstrate that the proposed method requires a lower annotation budget than
existing active learning methods to reach the same level of accuracy.
| no_new_dataset | 0.940898 |
2303.14435 | Zhiwen Yan | Zhiwen Yan, Chen Li, Gim Hee Lee | NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects | CVPR 2023 | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic Neural Radiance Field (NeRF) is a powerful algorithm capable of
rendering photo-realistic novel view images from a monocular RGB video of a
dynamic scene. Although it warps moving points across frames from the
observation spaces to a common canonical space for rendering, dynamic NeRF does
not model the change of the reflected color during the warping. As a result,
this approach often fails drastically on challenging specular objects in
motion. We address this limitation by reformulating the neural radiance field
function to be conditioned on surface position and orientation in the
observation space. This allows the specular surface at different poses to keep
the different reflected colors when mapped to the common canonical space.
Additionally, we add the mask of moving objects to guide the deformation field.
As the specular surface changes color during motion, the mask mitigates the
problem of failure to find temporal correspondences with only RGB supervision.
We evaluate our model based on the novel view synthesis quality with a
self-collected dataset of different moving specular objects in realistic
environments. The experimental results demonstrate that our method
significantly improves the reconstruction quality of moving specular objects
from monocular RGB videos compared to the existing NeRF models. Our code and
data are available at the project website https://github.com/JokerYan/NeRF-DS.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 11:03:53 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yan",
"Zhiwen",
""
],
[
"Li",
"Chen",
""
],
[
"Lee",
"Gim Hee",
""
]
] | TITLE: NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects
ABSTRACT: Dynamic Neural Radiance Field (NeRF) is a powerful algorithm capable of
rendering photo-realistic novel view images from a monocular RGB video of a
dynamic scene. Although it warps moving points across frames from the
observation spaces to a common canonical space for rendering, dynamic NeRF does
not model the change of the reflected color during the warping. As a result,
this approach often fails drastically on challenging specular objects in
motion. We address this limitation by reformulating the neural radiance field
function to be conditioned on surface position and orientation in the
observation space. This allows the specular surface at different poses to keep
the different reflected colors when mapped to the common canonical space.
Additionally, we add the mask of moving objects to guide the deformation field.
As the specular surface changes color during motion, the mask mitigates the
problem of failure to find temporal correspondences with only RGB supervision.
We evaluate our model based on the novel view synthesis quality with a
self-collected dataset of different moving specular objects in realistic
environments. The experimental results demonstrate that our method
significantly improves the reconstruction quality of moving specular objects
from monocular RGB videos compared to the existing NeRF models. Our code and
data are available at the project website https://github.com/JokerYan/NeRF-DS.
| new_dataset | 0.974043 |
2303.14453 | Yilun Jin | Yilun Jin, Yang Liu, Kai Chen, Qiang Yang | Federated Learning without Full Labels: A Survey | This survey is an extended and contemporary version of
arXiv:2002.11545 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Data privacy has become an increasingly important concern in real-world big
data applications such as machine learning. To address the problem, federated
learning (FL) has been a promising solution to building effective machine
learning models from decentralized and private data. Existing federated
learning algorithms mainly tackle the supervised learning problem, where data
are assumed to be fully labeled. However, in practice, fully labeled data is
often hard to obtain, as the participants may not have sufficient domain
expertise, or they lack the motivation and tools to label data. Therefore, the
problem of federated learning without full labels is important in real-world FL
applications. In this paper, we discuss how the problem can be solved with
machine learning techniques that leverage unlabeled data. We present a survey
of methods that combine FL with semi-supervised learning, self-supervised
learning, and transfer learning methods. We also summarize the datasets used to
evaluate FL methods without full labels. Finally, we highlight future
directions in the context of FL without full labels.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 12:13:31 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jin",
"Yilun",
""
],
[
"Liu",
"Yang",
""
],
[
"Chen",
"Kai",
""
],
[
"Yang",
"Qiang",
""
]
] | TITLE: Federated Learning without Full Labels: A Survey
ABSTRACT: Data privacy has become an increasingly important concern in real-world big
data applications such as machine learning. To address the problem, federated
learning (FL) has been a promising solution to building effective machine
learning models from decentralized and private data. Existing federated
learning algorithms mainly tackle the supervised learning problem, where data
are assumed to be fully labeled. However, in practice, fully labeled data is
often hard to obtain, as the participants may not have sufficient domain
expertise, or they lack the motivation and tools to label data. Therefore, the
problem of federated learning without full labels is important in real-world FL
applications. In this paper, we discuss how the problem can be solved with
machine learning techniques that leverage unlabeled data. We present a survey
of methods that combine FL with semi-supervised learning, self-supervised
learning, and transfer learning methods. We also summarize the datasets used to
evaluate FL methods without full labels. Finally, we highlight future
directions in the context of FL without full labels.
| no_new_dataset | 0.948537 |
2303.14457 | Tianxiang Ren | Tianxiang Ren, Jubo Yu, Shihui Guo, Ying Ma, Yutao Ouyang, Zijiao
Zeng, Yazhan Zhang, Yipeng Qin | Diverse Motion In-betweening with Dual Posture Stitching | 10 pages, 5 figures | null | null | null | cs.CV cs.AI cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In-betweening is a technique for generating transitions given initial and
target character states. The majority of existing works require multiple (often
$>$10) frames as input, which are not always accessible. Our work deals with a
focused yet challenging problem: to generate the transition when given exactly
two frames (only the first and last). To cope with this challenging scenario,
we implement our bi-directional scheme which generates forward and backward
transitions from the start and end frames with two adversarial autoregressive
networks, and stitches them in the middle of the transition where there is no
strict ground truth. The autoregressive networks based on conditional
variational autoencoders (CVAE) are optimized by searching for a pair of
optimal latent codes that minimize a novel stitching loss between their
outputs. Results show that our method achieves higher motion quality and more
diverse results than existing methods on both the LaFAN1 and Human3.6m
datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 12:36:46 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ren",
"Tianxiang",
""
],
[
"Yu",
"Jubo",
""
],
[
"Guo",
"Shihui",
""
],
[
"Ma",
"Ying",
""
],
[
"Ouyang",
"Yutao",
""
],
[
"Zeng",
"Zijiao",
""
],
[
"Zhang",
"Yazhan",
""
],
[
"Qin",
"Yipeng",
""
]
] | TITLE: Diverse Motion In-betweening with Dual Posture Stitching
ABSTRACT: In-betweening is a technique for generating transitions given initial and
target character states. The majority of existing works require multiple (often
$>$10) frames as input, which are not always accessible. Our work deals with a
focused yet challenging problem: to generate the transition when given exactly
two frames (only the first and last). To cope with this challenging scenario,
we implement our bi-directional scheme which generates forward and backward
transitions from the start and end frames with two adversarial autoregressive
networks, and stitches them in the middle of the transition where there is no
strict ground truth. The autoregressive networks based on conditional
variational autoencoders (CVAE) are optimized by searching for a pair of
optimal latent codes that minimize a novel stitching loss between their
outputs. Results show that our method achieves higher motion quality and more
diverse results than existing methods on both the LaFAN1 and Human3.6m
datasets.
| no_new_dataset | 0.944944 |
2303.14460 | Zeming Wei | Zeming Wei, Yifei Wang, Yiwen Guo, Yisen Wang | CFA: Class-wise Calibrated Fair Adversarial Training | CVPR 2023 | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial training has been widely acknowledged as the most effective
method to improve the adversarial robustness against adversarial examples for
Deep Neural Networks (DNNs). So far, most existing works focus on enhancing the
overall model robustness, treating each class equally in both the training and
testing phases. Although revealing the disparity in robustness among classes,
few works try to make adversarial training fair at the class level without
sacrificing overall robustness. In this paper, we are the first to
theoretically and empirically investigate the preference of different classes
for adversarial configurations, including perturbation margin, regularization,
and weight averaging. Motivated by this, we further propose a
\textbf{C}lass-wise calibrated \textbf{F}air \textbf{A}dversarial training
framework, named CFA, which customizes specific training configurations for
each class automatically. Experiments on benchmark datasets demonstrate that
our proposed CFA can improve both overall robustness and fairness notably over
other state-of-the-art methods. Code is available at
\url{https://github.com/PKU-ML/CFA}.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 13:05:16 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wei",
"Zeming",
""
],
[
"Wang",
"Yifei",
""
],
[
"Guo",
"Yiwen",
""
],
[
"Wang",
"Yisen",
""
]
] | TITLE: CFA: Class-wise Calibrated Fair Adversarial Training
ABSTRACT: Adversarial training has been widely acknowledged as the most effective
method to improve the adversarial robustness against adversarial examples for
Deep Neural Networks (DNNs). So far, most existing works focus on enhancing the
overall model robustness, treating each class equally in both the training and
testing phases. Although revealing the disparity in robustness among classes,
few works try to make adversarial training fair at the class level without
sacrificing overall robustness. In this paper, we are the first to
theoretically and empirically investigate the preference of different classes
for adversarial configurations, including perturbation margin, regularization,
and weight averaging. Motivated by this, we further propose a
\textbf{C}lass-wise calibrated \textbf{F}air \textbf{A}dversarial training
framework, named CFA, which customizes specific training configurations for
each class automatically. Experiments on benchmark datasets demonstrate that
our proposed CFA can improve both overall robustness and fairness notably over
other state-of-the-art methods. Code is available at
\url{https://github.com/PKU-ML/CFA}.
| no_new_dataset | 0.95096 |
2303.14471 | Kartik Teotia | Kartik Teotia, Mallikarjun B R, Xingang Pan, Hyeongwoo Kim, Pablo
Garrido, Mohamed Elgharib, Christian Theobalt | HQ3DAvatar: High Quality Controllable 3D Head Avatar | 16 Pages, 15 Figures. Project page:
https://vcai.mpi-inf.mpg.de/projects/HQ3DAvatar/ | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by/4.0/ | Multi-view volumetric rendering techniques have recently shown great
potential in modeling and synthesizing high-quality head avatars. A common
approach to capture full head dynamic performances is to track the underlying
geometry using a mesh-based template or 3D cube-based graphics primitives.
While these model-based approaches achieve promising results, they often fail
to learn complex geometric details such as the mouth interior, hair, and
topological changes over time. This paper presents a novel approach to building
highly photorealistic digital head avatars. Our method learns a canonical space
via an implicit function parameterized by a neural network. It leverages
multiresolution hash encoding in the learned feature space, allowing for
high-quality, faster training and high-resolution rendering. At test time, our
method is driven by a monocular RGB video. Here, an image encoder extracts
face-specific features that also condition the learnable canonical space. This
encourages deformation-dependent texture variations during training. We also
propose a novel optical flow based loss that ensures correspondences in the
learned canonical space, thus encouraging artifact-free and temporally
consistent renderings. We show results on challenging facial expressions and
show free-viewpoint renderings at interactive real-time rates for medium image
resolutions. Our method outperforms all existing approaches, both visually and
numerically. We will release our multiple-identity dataset to encourage further
research. Our Project page is available at:
https://vcai.mpi-inf.mpg.de/projects/HQ3DAvatar/
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 13:56:33 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Teotia",
"Kartik",
""
],
[
"R",
"Mallikarjun B",
""
],
[
"Pan",
"Xingang",
""
],
[
"Kim",
"Hyeongwoo",
""
],
[
"Garrido",
"Pablo",
""
],
[
"Elgharib",
"Mohamed",
""
],
[
"Theobalt",
"Christian",
""
]
] | TITLE: HQ3DAvatar: High Quality Controllable 3D Head Avatar
ABSTRACT: Multi-view volumetric rendering techniques have recently shown great
potential in modeling and synthesizing high-quality head avatars. A common
approach to capture full head dynamic performances is to track the underlying
geometry using a mesh-based template or 3D cube-based graphics primitives.
While these model-based approaches achieve promising results, they often fail
to learn complex geometric details such as the mouth interior, hair, and
topological changes over time. This paper presents a novel approach to building
highly photorealistic digital head avatars. Our method learns a canonical space
via an implicit function parameterized by a neural network. It leverages
multiresolution hash encoding in the learned feature space, allowing for
high-quality, faster training and high-resolution rendering. At test time, our
method is driven by a monocular RGB video. Here, an image encoder extracts
face-specific features that also condition the learnable canonical space. This
encourages deformation-dependent texture variations during training. We also
propose a novel optical flow based loss that ensures correspondences in the
learned canonical space, thus encouraging artifact-free and temporally
consistent renderings. We show results on challenging facial expressions and
show free-viewpoint renderings at interactive real-time rates for medium image
resolutions. Our method outperforms all existing approaches, both visually and
numerically. We will release our multiple-identity dataset to encourage further
research. Our Project page is available at:
https://vcai.mpi-inf.mpg.de/projects/HQ3DAvatar/
| new_dataset | 0.961207 |
2303.14478 | Yu Chen | Yu Chen, Gim Hee Lee | DBARF: Deep Bundle-Adjusting Generalizable Neural Radiance Fields | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent works such as BARF and GARF can bundle adjust camera poses with neural
radiance fields (NeRF) which is based on coordinate-MLPs. Despite the
impressive results, these methods cannot be applied to Generalizable NeRFs
(GeNeRFs) which require image feature extractions that are often based on more
complicated 3D CNN or transformer architectures. In this work, we first analyze
the difficulties of jointly optimizing camera poses with GeNeRFs, and then
further propose our DBARF to tackle these issues. Our DBARF which bundle
adjusts camera poses by taking a cost feature map as an implicit cost function
can be jointly trained with GeNeRFs in a self-supervised manner. Unlike BARF
and its follow-up works, which can only be applied to per-scene optimized NeRFs
and need accurate initial camera poses with the exception of forward-facing
scenes, our method can generalize across scenes and does not require any good
initialization. Experiments show the effectiveness and generalization ability
of our DBARF when evaluated on real-world datasets. Our code is available at
\url{https://aibluefisher.github.io/dbarf}.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 14:18:30 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Yu",
""
],
[
"Lee",
"Gim Hee",
""
]
] | TITLE: DBARF: Deep Bundle-Adjusting Generalizable Neural Radiance Fields
ABSTRACT: Recent works such as BARF and GARF can bundle adjust camera poses with neural
radiance fields (NeRF) which is based on coordinate-MLPs. Despite the
impressive results, these methods cannot be applied to Generalizable NeRFs
(GeNeRFs) which require image feature extractions that are often based on more
complicated 3D CNN or transformer architectures. In this work, we first analyze
the difficulties of jointly optimizing camera poses with GeNeRFs, and then
further propose our DBARF to tackle these issues. Our DBARF which bundle
adjusts camera poses by taking a cost feature map as an implicit cost function
can be jointly trained with GeNeRFs in a self-supervised manner. Unlike BARF
and its follow-up works, which can only be applied to per-scene optimized NeRFs
and need accurate initial camera poses with the exception of forward-facing
scenes, our method can generalize across scenes and does not require any good
initialization. Experiments show the effectiveness and generalization ability
of our DBARF when evaluated on real-world datasets. Our code is available at
\url{https://aibluefisher.github.io/dbarf}.
| no_new_dataset | 0.950365 |
2303.14479 | Caner Ozer | Caner Ozer, Arda Guler, Aysel Turkvatan Cansever, Ilkay Oksuz | Explainable Image Quality Assessment for Medical Imaging | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | Medical image quality assessment is an important aspect of image acquisition,
as poor-quality images may lead to misdiagnosis. Manual labelling of image
quality is a tedious task for population studies and can lead to misleading
results. While much research has been done on automated analysis of image
quality to address this issue, relatively little work has been done to explain
the methodologies. In this work, we propose an explainable image quality
assessment system and validate our idea on two different objectives which are
foreign object detection on Chest X-Rays (Object-CXR) and Left Ventricular
Outflow Tract (LVOT) detection on Cardiac Magnetic Resonance (CMR) volumes. We
apply a variety of techniques to measure the faithfulness of the saliency
detectors, and our explainable pipeline relies on NormGrad, an algorithm which
can efficiently localise image quality issues with saliency maps of the
classifier. We compare NormGrad with a range of saliency detection methods and
illustrate its superior performance as a result of applying these methodologies
for measuring the faithfulness of the saliency detectors. We see that NormGrad
has significant gains over other saliency detectors by reaching a repeated
Pointing Game score of 0.853 for Object-CXR and 0.611 for LVOT datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 14:18:39 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ozer",
"Caner",
""
],
[
"Guler",
"Arda",
""
],
[
"Cansever",
"Aysel Turkvatan",
""
],
[
"Oksuz",
"Ilkay",
""
]
] | TITLE: Explainable Image Quality Assessment for Medical Imaging
ABSTRACT: Medical image quality assessment is an important aspect of image acquisition,
as poor-quality images may lead to misdiagnosis. Manual labelling of image
quality is a tedious task for population studies and can lead to misleading
results. While much research has been done on automated analysis of image
quality to address this issue, relatively little work has been done to explain
the methodologies. In this work, we propose an explainable image quality
assessment system and validate our idea on two different objectives which are
foreign object detection on Chest X-Rays (Object-CXR) and Left Ventricular
Outflow Tract (LVOT) detection on Cardiac Magnetic Resonance (CMR) volumes. We
apply a variety of techniques to measure the faithfulness of the saliency
detectors, and our explainable pipeline relies on NormGrad, an algorithm which
can efficiently localise image quality issues with saliency maps of the
classifier. We compare NormGrad with a range of saliency detection methods and
illustrate its superior performance as a result of applying these methodologies
for measuring the faithfulness of the saliency detectors. We see that NormGrad
has significant gains over other saliency detectors by reaching a repeated
Pointing Game score of 0.853 for Object-CXR and 0.611 for LVOT datasets.
| no_new_dataset | 0.94801 |
2303.14480 | Zhouhong Gu | Zhouhong Gu, Sihang Jiang, Jingping Liu, Yanghua Xiao, Hongwei Feng,
Zhixu Li, Jiaqing Liang, Jian Zhong | GANTEE: Generative Adversatial Network for Taxonomy Entering Evaluation | Accepted by AAAI 2023 | null | null | null | cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Taxonomy is formulated as directed acyclic concepts graphs or trees that
support many downstream tasks. Many new coming concepts need to be added to an
existing taxonomy. The traditional taxonomy expansion task aims only at finding
the best position for new coming concepts in the existing taxonomy. However,
they have two drawbacks when being applied to the real-scenarios. The previous
methods suffer from low-efficiency since they waste much time when most of the
new coming concepts are indeed noisy concepts. They also suffer from
low-effectiveness since they collect training samples only from the existing
taxonomy, which limits the ability of the model to mine more hypernym-hyponym
relationships among real concepts. This paper proposes a pluggable framework
called Generative Adversarial Network for Taxonomy Entering Evaluation (GANTEE)
to alleviate these drawbacks. A generative adversarial network is designed in
this framework by discriminative models to alleviate the first drawback and the
generative model to alleviate the second drawback. Two discriminators are used
in GANTEE to provide long-term and short-term rewards, respectively. Moreover,
to further improve the efficiency, pre-trained language models are used to
retrieve the representation of the concepts quickly. The experiments on three
real-world large-scale datasets with two different languages show that GANTEE
improves the performance of the existing taxonomy expansion methods in both
effectiveness and efficiency.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 14:24:50 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Gu",
"Zhouhong",
""
],
[
"Jiang",
"Sihang",
""
],
[
"Liu",
"Jingping",
""
],
[
"Xiao",
"Yanghua",
""
],
[
"Feng",
"Hongwei",
""
],
[
"Li",
"Zhixu",
""
],
[
"Liang",
"Jiaqing",
""
],
[
"Zhong",
"Jian",
""
]
] | TITLE: GANTEE: Generative Adversatial Network for Taxonomy Entering Evaluation
ABSTRACT: Taxonomy is formulated as directed acyclic concepts graphs or trees that
support many downstream tasks. Many new coming concepts need to be added to an
existing taxonomy. The traditional taxonomy expansion task aims only at finding
the best position for new coming concepts in the existing taxonomy. However,
they have two drawbacks when being applied to the real-scenarios. The previous
methods suffer from low-efficiency since they waste much time when most of the
new coming concepts are indeed noisy concepts. They also suffer from
low-effectiveness since they collect training samples only from the existing
taxonomy, which limits the ability of the model to mine more hypernym-hyponym
relationships among real concepts. This paper proposes a pluggable framework
called Generative Adversarial Network for Taxonomy Entering Evaluation (GANTEE)
to alleviate these drawbacks. A generative adversarial network is designed in
this framework by discriminative models to alleviate the first drawback and the
generative model to alleviate the second drawback. Two discriminators are used
in GANTEE to provide long-term and short-term rewards, respectively. Moreover,
to further improve the efficiency, pre-trained language models are used to
retrieve the representation of the concepts quickly. The experiments on three
real-world large-scale datasets with two different languages show that GANTEE
improves the performance of the existing taxonomy expansion methods in both
effectiveness and efficiency.
| no_new_dataset | 0.954137 |
2303.14481 | Yukang Zhang | Yukang Zhang, Hanzi Wang | Diverse Embedding Expansion Network and Low-Light Cross-Modality
Benchmark for Visible-Infrared Person Re-identification | Accepted by CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | For the visible-infrared person re-identification (VIReID) task, one of the
major challenges is the modality gaps between visible (VIS) and infrared (IR)
images. However, the training samples are usually limited, while the modality
gaps are too large, which leads that the existing methods cannot effectively
mine diverse cross-modality clues. To handle this limitation, we propose a
novel augmentation network in the embedding space, called diverse embedding
expansion network (DEEN). The proposed DEEN can effectively generate diverse
embeddings to learn the informative feature representations and reduce the
modality discrepancy between the VIS and IR images. Moreover, the VIReID model
may be seriously affected by drastic illumination changes, while all the
existing VIReID datasets are captured under sufficient illumination without
significant light changes. Thus, we provide a low-light cross-modality (LLCM)
dataset, which contains 46,767 bounding boxes of 1,064 identities captured by 9
RGB/IR cameras. Extensive experiments on the SYSU-MM01, RegDB and LLCM datasets
show the superiority of the proposed DEEN over several other state-of-the-art
methods. The code and dataset are released at: https://github.com/ZYK100/LLCM
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 14:24:56 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Yukang",
""
],
[
"Wang",
"Hanzi",
""
]
] | TITLE: Diverse Embedding Expansion Network and Low-Light Cross-Modality
Benchmark for Visible-Infrared Person Re-identification
ABSTRACT: For the visible-infrared person re-identification (VIReID) task, one of the
major challenges is the modality gaps between visible (VIS) and infrared (IR)
images. However, the training samples are usually limited, while the modality
gaps are too large, which leads that the existing methods cannot effectively
mine diverse cross-modality clues. To handle this limitation, we propose a
novel augmentation network in the embedding space, called diverse embedding
expansion network (DEEN). The proposed DEEN can effectively generate diverse
embeddings to learn the informative feature representations and reduce the
modality discrepancy between the VIS and IR images. Moreover, the VIReID model
may be seriously affected by drastic illumination changes, while all the
existing VIReID datasets are captured under sufficient illumination without
significant light changes. Thus, we provide a low-light cross-modality (LLCM)
dataset, which contains 46,767 bounding boxes of 1,064 identities captured by 9
RGB/IR cameras. Extensive experiments on the SYSU-MM01, RegDB and LLCM datasets
show the superiority of the proposed DEEN over several other state-of-the-art
methods. The code and dataset are released at: https://github.com/ZYK100/LLCM
| new_dataset | 0.966442 |
2303.14488 | Bowei Du | Bowei Du, Yecheng Huang, Jiaxin Chen, Di Huang | Adaptive Sparse Convolutional Networks with Global Context Enhancement
for Faster Object Detection on Drone Images | Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Object detection on drone images with low-latency is an important but
challenging task on the resource-constrained unmanned aerial vehicle (UAV)
platform. This paper investigates optimizing the detection head based on the
sparse convolution, which proves effective in balancing the accuracy and
efficiency. Nevertheless, it suffers from inadequate integration of contextual
information of tiny objects as well as clumsy control of the mask ratio in the
presence of foreground with varying scales. To address the issues above, we
propose a novel global context-enhanced adaptive sparse convolutional network
(CEASC). It first develops a context-enhanced group normalization (CE-GN)
layer, by replacing the statistics based on sparsely sampled features with the
global contextual ones, and then designs an adaptive multi-layer masking
strategy to generate optimal mask ratios at distinct scales for compact
foreground coverage, promoting both the accuracy and efficiency. Extensive
experimental results on two major benchmarks, i.e. VisDrone and UAVDT,
demonstrate that CEASC remarkably reduces the GFLOPs and accelerates the
inference procedure when plugging into the typical state-of-the-art detection
frameworks (e.g. RetinaNet and GFL V1) with competitive performance. Code is
available at https://github.com/Cuogeihong/CEASC.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 14:42:50 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Du",
"Bowei",
""
],
[
"Huang",
"Yecheng",
""
],
[
"Chen",
"Jiaxin",
""
],
[
"Huang",
"Di",
""
]
] | TITLE: Adaptive Sparse Convolutional Networks with Global Context Enhancement
for Faster Object Detection on Drone Images
ABSTRACT: Object detection on drone images with low-latency is an important but
challenging task on the resource-constrained unmanned aerial vehicle (UAV)
platform. This paper investigates optimizing the detection head based on the
sparse convolution, which proves effective in balancing the accuracy and
efficiency. Nevertheless, it suffers from inadequate integration of contextual
information of tiny objects as well as clumsy control of the mask ratio in the
presence of foreground with varying scales. To address the issues above, we
propose a novel global context-enhanced adaptive sparse convolutional network
(CEASC). It first develops a context-enhanced group normalization (CE-GN)
layer, by replacing the statistics based on sparsely sampled features with the
global contextual ones, and then designs an adaptive multi-layer masking
strategy to generate optimal mask ratios at distinct scales for compact
foreground coverage, promoting both the accuracy and efficiency. Extensive
experimental results on two major benchmarks, i.e. VisDrone and UAVDT,
demonstrate that CEASC remarkably reduces the GFLOPs and accelerates the
inference procedure when plugging into the typical state-of-the-art detection
frameworks (e.g. RetinaNet and GFL V1) with competitive performance. Code is
available at https://github.com/Cuogeihong/CEASC.
| no_new_dataset | 0.948489 |
2303.14498 | Wenqiang Xu | Wenqiang Xu, Zhenjun Yu, Han Xue, Ruolin Ye, Siqiong Yao, Cewu Lu | Visual-Tactile Sensing for In-Hand Object Reconstruction | Accepted in CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Tactile sensing is one of the modalities humans rely on heavily to perceive
the world. Working with vision, this modality refines local geometry structure,
measures deformation at the contact area, and indicates the hand-object contact
state.
With the availability of open-source tactile sensors such as DIGIT, research
on visual-tactile learning is becoming more accessible and reproducible.
Leveraging this tactile sensor, we propose a novel visual-tactile in-hand
object reconstruction framework \textbf{VTacO}, and extend it to
\textbf{VTacOH} for hand-object reconstruction. Since our method can support
both rigid and deformable object reconstruction, no existing benchmarks are
proper for the goal. We propose a simulation environment, VT-Sim, which
supports generating hand-object interaction for both rigid and deformable
objects. With VT-Sim, we generate a large-scale training dataset and evaluate
our method on it. Extensive experiments demonstrate that our proposed method
can outperform the previous baseline methods qualitatively and quantitatively.
Finally, we directly apply our model trained in simulation to various
real-world test cases, which display qualitative results.
Codes, models, simulation environment, and datasets are available at
\url{https://sites.google.com/view/vtaco/}.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 15:16:31 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Xu",
"Wenqiang",
""
],
[
"Yu",
"Zhenjun",
""
],
[
"Xue",
"Han",
""
],
[
"Ye",
"Ruolin",
""
],
[
"Yao",
"Siqiong",
""
],
[
"Lu",
"Cewu",
""
]
] | TITLE: Visual-Tactile Sensing for In-Hand Object Reconstruction
ABSTRACT: Tactile sensing is one of the modalities humans rely on heavily to perceive
the world. Working with vision, this modality refines local geometry structure,
measures deformation at the contact area, and indicates the hand-object contact
state.
With the availability of open-source tactile sensors such as DIGIT, research
on visual-tactile learning is becoming more accessible and reproducible.
Leveraging this tactile sensor, we propose a novel visual-tactile in-hand
object reconstruction framework \textbf{VTacO}, and extend it to
\textbf{VTacOH} for hand-object reconstruction. Since our method can support
both rigid and deformable object reconstruction, no existing benchmarks are
proper for the goal. We propose a simulation environment, VT-Sim, which
supports generating hand-object interaction for both rigid and deformable
objects. With VT-Sim, we generate a large-scale training dataset and evaluate
our method on it. Extensive experiments demonstrate that our proposed method
can outperform the previous baseline methods qualitatively and quantitatively.
Finally, we directly apply our model trained in simulation to various
real-world test cases, which display qualitative results.
Codes, models, simulation environment, and datasets are available at
\url{https://sites.google.com/view/vtaco/}.
| no_new_dataset | 0.556595 |
2303.14505 | Chao Chen | Chao Chen, Yu-Shen Liu, Zhizhong Han | Unsupervised Inference of Signed Distance Functions from Single Sparse
Point Clouds without Learning Priors | 12pages,12figures. Accepted by CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is vital to infer signed distance functions (SDFs) from 3D point clouds.
The latest methods rely on generalizing the priors learned from large scale
supervision. However, the learned priors do not generalize well to various
geometric variations that are unseen during training, especially for extremely
sparse point clouds. To resolve this issue, we present a neural network to
directly infer SDFs from single sparse point clouds without using signed
distance supervision, learned priors or even normals. Our insight here is to
learn surface parameterization and SDFs inference in an end-to-end manner. To
make up the sparsity, we leverage parameterized surfaces as a coarse surface
sampler to provide many coarse surface estimations in training iterations,
according to which we mine supervision and our thin plate splines (TPS) based
network infers SDFs as smooth functions in a statistical way. Our method
significantly improves the generalization ability and accuracy in unseen point
clouds. Our experimental results show our advantages over the state-of-the-art
methods in surface reconstruction for sparse point clouds under synthetic
datasets and real scans.The code is available at
\url{https://github.com/chenchao15/NeuralTPS}.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 15:56:50 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Chao",
""
],
[
"Liu",
"Yu-Shen",
""
],
[
"Han",
"Zhizhong",
""
]
] | TITLE: Unsupervised Inference of Signed Distance Functions from Single Sparse
Point Clouds without Learning Priors
ABSTRACT: It is vital to infer signed distance functions (SDFs) from 3D point clouds.
The latest methods rely on generalizing the priors learned from large scale
supervision. However, the learned priors do not generalize well to various
geometric variations that are unseen during training, especially for extremely
sparse point clouds. To resolve this issue, we present a neural network to
directly infer SDFs from single sparse point clouds without using signed
distance supervision, learned priors or even normals. Our insight here is to
learn surface parameterization and SDFs inference in an end-to-end manner. To
make up the sparsity, we leverage parameterized surfaces as a coarse surface
sampler to provide many coarse surface estimations in training iterations,
according to which we mine supervision and our thin plate splines (TPS) based
network infers SDFs as smooth functions in a statistical way. Our method
significantly improves the generalization ability and accuracy in unseen point
clouds. Our experimental results show our advantages over the state-of-the-art
methods in surface reconstruction for sparse point clouds under synthetic
datasets and real scans.The code is available at
\url{https://github.com/chenchao15/NeuralTPS}.
| no_new_dataset | 0.952309 |
2303.14510 | Wensheng Gan | Shan Huang, Wensheng Gan, Jinbao Miao, Xuming Han, Philippe
Fournier-Viger | Targeted Mining of Top-k High Utility Itemsets | Preprint. 5 figures, 5 tables | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Finding high-importance patterns in data is an emerging data mining task
known as High-utility itemset mining (HUIM). Given a minimum utility threshold,
a HUIM algorithm extracts all the high-utility itemsets (HUIs) whose utility
values are not less than the threshold. This can reveal a wealth of useful
information, but the precise needs of users are not well taken into account. In
particular, users often want to focus on patterns that have some specific items
rather than find all patterns. To overcome that difficulty, targeted mining has
emerged, focusing on user preferences, but only preliminary work has been
conducted. For example, the targeted high-utility itemset querying algorithm
(TargetUM) was proposed, which uses a lexicographic tree to query itemsets
containing a target pattern. However, selecting the minimum utility threshold
is difficult when the user is not familiar with the processed database. As a
solution, this paper formulates the task of targeted mining of the top-k
high-utility itemsets and proposes an efficient algorithm called TMKU based on
the TargetUM algorithm to discover the top-k target high-utility itemsets
(top-k THUIs). At the same time, several pruning strategies are used to reduce
memory consumption and execution time. Extensive experiments show that the
proposed TMKU algorithm has good performance on real and synthetic datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 16:19:39 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Huang",
"Shan",
""
],
[
"Gan",
"Wensheng",
""
],
[
"Miao",
"Jinbao",
""
],
[
"Han",
"Xuming",
""
],
[
"Fournier-Viger",
"Philippe",
""
]
] | TITLE: Targeted Mining of Top-k High Utility Itemsets
ABSTRACT: Finding high-importance patterns in data is an emerging data mining task
known as High-utility itemset mining (HUIM). Given a minimum utility threshold,
a HUIM algorithm extracts all the high-utility itemsets (HUIs) whose utility
values are not less than the threshold. This can reveal a wealth of useful
information, but the precise needs of users are not well taken into account. In
particular, users often want to focus on patterns that have some specific items
rather than find all patterns. To overcome that difficulty, targeted mining has
emerged, focusing on user preferences, but only preliminary work has been
conducted. For example, the targeted high-utility itemset querying algorithm
(TargetUM) was proposed, which uses a lexicographic tree to query itemsets
containing a target pattern. However, selecting the minimum utility threshold
is difficult when the user is not familiar with the processed database. As a
solution, this paper formulates the task of targeted mining of the top-k
high-utility itemsets and proposes an efficient algorithm called TMKU based on
the TargetUM algorithm to discover the top-k target high-utility itemsets
(top-k THUIs). At the same time, several pruning strategies are used to reduce
memory consumption and execution time. Extensive experiments show that the
proposed TMKU algorithm has good performance on real and synthetic datasets.
| no_new_dataset | 0.949949 |
2303.14517 | Surya Mahadi | Made Raharja Surya Mahadi and Nugraha Priya Utama | Indonesian Text-to-Image Synthesis with Sentence-BERT and FastGAN | 11 pages, 3 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Currently, text-to-image synthesis uses text encoder and image generator
architecture. Research on this topic is challenging. This is because of the
domain gap between natural language and vision. Nowadays, most research on this
topic only focuses on producing a photo-realistic image, but the other domain,
in this case, is the language, which is less concentrated. A lot of the current
research uses English as the input text. Besides, there are many languages
around the world. Bahasa Indonesia, as the official language of Indonesia, is
quite popular. This language has been taught in Philipines, Australia, and
Japan. Translating or recreating a new dataset into another language with good
quality will cost a lot. Research on this domain is necessary because we need
to examine how the image generator performs in other languages besides
generating photo-realistic images. To achieve this, we translate the CUB
dataset into Bahasa using google translate and manually by humans. We use
Sentence BERT as the text encoder and FastGAN as the image generator. FastGAN
uses lots of skip excitation modules and auto-encoder to generate an image with
resolution 512x512x3, which is twice as bigger as the current state-of-the-art
model (Zhang, Xu, Li, Zhang, Wang, Huang and Metaxas, 2019). We also get 4.76
+- 0.43 and 46.401 on Inception Score and Fr\'echet inception distance,
respectively, and comparable with the current English text-to-image generation
models. The mean opinion score also gives as 3.22 out of 5, which means the
generated image is acceptable by humans. Link to source code:
https://github.com/share424/Indonesian-Text-to-Image-synthesis-with-Sentence-BERT-and-FastGAN
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 16:54:22 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Mahadi",
"Made Raharja Surya",
""
],
[
"Utama",
"Nugraha Priya",
""
]
] | TITLE: Indonesian Text-to-Image Synthesis with Sentence-BERT and FastGAN
ABSTRACT: Currently, text-to-image synthesis uses text encoder and image generator
architecture. Research on this topic is challenging. This is because of the
domain gap between natural language and vision. Nowadays, most research on this
topic only focuses on producing a photo-realistic image, but the other domain,
in this case, is the language, which is less concentrated. A lot of the current
research uses English as the input text. Besides, there are many languages
around the world. Bahasa Indonesia, as the official language of Indonesia, is
quite popular. This language has been taught in Philipines, Australia, and
Japan. Translating or recreating a new dataset into another language with good
quality will cost a lot. Research on this domain is necessary because we need
to examine how the image generator performs in other languages besides
generating photo-realistic images. To achieve this, we translate the CUB
dataset into Bahasa using google translate and manually by humans. We use
Sentence BERT as the text encoder and FastGAN as the image generator. FastGAN
uses lots of skip excitation modules and auto-encoder to generate an image with
resolution 512x512x3, which is twice as bigger as the current state-of-the-art
model (Zhang, Xu, Li, Zhang, Wang, Huang and Metaxas, 2019). We also get 4.76
+- 0.43 and 46.401 on Inception Score and Fr\'echet inception distance,
respectively, and comparable with the current English text-to-image generation
models. The mean opinion score also gives as 3.22 out of 5, which means the
generated image is acceptable by humans. Link to source code:
https://github.com/share424/Indonesian-Text-to-Image-synthesis-with-Sentence-BERT-and-FastGAN
| no_new_dataset | 0.945901 |
2303.14526 | Jue Wang | Jue Wang, Wentao Zhu, Pichao Wang, Xiang Yu, Linda Liu, Mohamed Omar,
Raffay Hamid | Selective Structured State-Spaces for Long-Form Video Understanding | Accepted by CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Effective modeling of complex spatiotemporal dependencies in long-form videos
remains an open problem. The recently proposed Structured State-Space Sequence
(S4) model with its linear complexity offers a promising direction in this
space. However, we demonstrate that treating all image-tokens equally as done
by S4 model can adversely affect its efficiency and accuracy. To address this
limitation, we present a novel Selective S4 (i.e., S5) model that employs a
lightweight mask generator to adaptively select informative image tokens
resulting in more efficient and accurate modeling of long-term spatiotemporal
dependencies in videos. Unlike previous mask-based token reduction methods used
in transformers, our S5 model avoids the dense self-attention calculation by
making use of the guidance of the momentum-updated S4 model. This enables our
model to efficiently discard less informative tokens and adapt to various
long-form video understanding tasks more effectively. However, as is the case
for most token reduction methods, the informative image tokens could be dropped
incorrectly. To improve the robustness and the temporal horizon of our model,
we propose a novel long-short masked contrastive learning (LSMCL) approach that
enables our model to predict longer temporal context using shorter input
videos. We present extensive comparative results using three challenging
long-form video understanding datasets (LVU, COIN and Breakfast), demonstrating
that our approach consistently outperforms the previous state-of-the-art S4
model by up to 9.6% accuracy while reducing its memory footprint by 23%.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 17:47:12 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Jue",
""
],
[
"Zhu",
"Wentao",
""
],
[
"Wang",
"Pichao",
""
],
[
"Yu",
"Xiang",
""
],
[
"Liu",
"Linda",
""
],
[
"Omar",
"Mohamed",
""
],
[
"Hamid",
"Raffay",
""
]
] | TITLE: Selective Structured State-Spaces for Long-Form Video Understanding
ABSTRACT: Effective modeling of complex spatiotemporal dependencies in long-form videos
remains an open problem. The recently proposed Structured State-Space Sequence
(S4) model with its linear complexity offers a promising direction in this
space. However, we demonstrate that treating all image-tokens equally as done
by S4 model can adversely affect its efficiency and accuracy. To address this
limitation, we present a novel Selective S4 (i.e., S5) model that employs a
lightweight mask generator to adaptively select informative image tokens
resulting in more efficient and accurate modeling of long-term spatiotemporal
dependencies in videos. Unlike previous mask-based token reduction methods used
in transformers, our S5 model avoids the dense self-attention calculation by
making use of the guidance of the momentum-updated S4 model. This enables our
model to efficiently discard less informative tokens and adapt to various
long-form video understanding tasks more effectively. However, as is the case
for most token reduction methods, the informative image tokens could be dropped
incorrectly. To improve the robustness and the temporal horizon of our model,
we propose a novel long-short masked contrastive learning (LSMCL) approach that
enables our model to predict longer temporal context using shorter input
videos. We present extensive comparative results using three challenging
long-form video understanding datasets (LVU, COIN and Breakfast), demonstrating
that our approach consistently outperforms the previous state-of-the-art S4
model by up to 9.6% accuracy while reducing its memory footprint by 23%.
| no_new_dataset | 0.950824 |
2303.14531 | Jingyang Zhang | Jingyang Zhang, Nathan Inkawhich, Randolph Linderman, Ryan Luley,
Yiran Chen, Hai Li | SIO: Synthetic In-Distribution Data Benefits Out-of-Distribution
Detection | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Building up reliable Out-of-Distribution (OOD) detectors is challenging,
often requiring the use of OOD data during training. In this work, we develop a
data-driven approach which is distinct and complementary to existing works:
Instead of using external OOD data, we fully exploit the internal
in-distribution (ID) training set by utilizing generative models to produce
additional synthetic ID images. The classifier is then trained using a novel
objective that computes weighted loss on real and synthetic ID samples
together. Our training framework, which is termed SIO, serves as a
"plug-and-play" technique that is designed to be compatible with existing and
future OOD detection algorithms, including the ones that leverage available OOD
training data. Our experiments on CIFAR-10, CIFAR-100, and ImageNet variants
demonstrate that SIO consistently improves the performance of nearly all
state-of-the-art (SOTA) OOD detection algorithms. For instance, on the
challenging CIFAR-10 v.s. CIFAR-100 detection problem, SIO improves the average
OOD detection AUROC of 18 existing methods from 86.25\% to 89.04\% and achieves
a new SOTA of 92.94\% according to the OpenOOD benchmark. Code is available at
https://github.com/zjysteven/SIO.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 18:34:34 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Jingyang",
""
],
[
"Inkawhich",
"Nathan",
""
],
[
"Linderman",
"Randolph",
""
],
[
"Luley",
"Ryan",
""
],
[
"Chen",
"Yiran",
""
],
[
"Li",
"Hai",
""
]
] | TITLE: SIO: Synthetic In-Distribution Data Benefits Out-of-Distribution
Detection
ABSTRACT: Building up reliable Out-of-Distribution (OOD) detectors is challenging,
often requiring the use of OOD data during training. In this work, we develop a
data-driven approach which is distinct and complementary to existing works:
Instead of using external OOD data, we fully exploit the internal
in-distribution (ID) training set by utilizing generative models to produce
additional synthetic ID images. The classifier is then trained using a novel
objective that computes weighted loss on real and synthetic ID samples
together. Our training framework, which is termed SIO, serves as a
"plug-and-play" technique that is designed to be compatible with existing and
future OOD detection algorithms, including the ones that leverage available OOD
training data. Our experiments on CIFAR-10, CIFAR-100, and ImageNet variants
demonstrate that SIO consistently improves the performance of nearly all
state-of-the-art (SOTA) OOD detection algorithms. For instance, on the
challenging CIFAR-10 v.s. CIFAR-100 detection problem, SIO improves the average
OOD detection AUROC of 18 existing methods from 86.25\% to 89.04\% and achieves
a new SOTA of 92.94\% according to the OpenOOD benchmark. Code is available at
https://github.com/zjysteven/SIO.
| no_new_dataset | 0.949012 |
2303.14543 | Yuzhou Chen | Yuzhou Chen, Yulia R. Gel | Topological Pooling on Graphs | AAAI 2023 | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Graph neural networks (GNNs) have demonstrated a significant success in
various graph learning tasks, from graph classification to anomaly detection.
There recently has emerged a number of approaches adopting a graph pooling
operation within GNNs, with a goal to preserve graph attributive and structural
features during the graph representation learning. However, most existing graph
pooling operations suffer from the limitations of relying on node-wise neighbor
weighting and embedding, which leads to insufficient encoding of rich
topological structures and node attributes exhibited by real-world networks. By
invoking the machinery of persistent homology and the concept of landmarks, we
propose a novel topological pooling layer and witness complex-based topological
embedding mechanism that allow us to systematically integrate hidden
topological information at both local and global levels. Specifically, we
design new learnable local and global topological representations Wit-TopoPool
which allow us to simultaneously extract rich discriminative topological
information from graphs. Experiments on 11 diverse benchmark datasets against
18 baseline models in conjunction with graph classification tasks indicate that
Wit-TopoPool significantly outperforms all competitors across all datasets.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 19:30:46 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Yuzhou",
""
],
[
"Gel",
"Yulia R.",
""
]
] | TITLE: Topological Pooling on Graphs
ABSTRACT: Graph neural networks (GNNs) have demonstrated a significant success in
various graph learning tasks, from graph classification to anomaly detection.
There recently has emerged a number of approaches adopting a graph pooling
operation within GNNs, with a goal to preserve graph attributive and structural
features during the graph representation learning. However, most existing graph
pooling operations suffer from the limitations of relying on node-wise neighbor
weighting and embedding, which leads to insufficient encoding of rich
topological structures and node attributes exhibited by real-world networks. By
invoking the machinery of persistent homology and the concept of landmarks, we
propose a novel topological pooling layer and witness complex-based topological
embedding mechanism that allow us to systematically integrate hidden
topological information at both local and global levels. Specifically, we
design new learnable local and global topological representations Wit-TopoPool
which allow us to simultaneously extract rich discriminative topological
information from graphs. Experiments on 11 diverse benchmark datasets against
18 baseline models in conjunction with graph classification tasks indicate that
Wit-TopoPool significantly outperforms all competitors across all datasets.
| no_new_dataset | 0.950041 |
2303.14552 | Maciej Sypetkowski | Maciej Sypetkowski | Spatial Latent Representations in Generative Adversarial Networks for
Image Generation | null | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the majority of GAN architectures, the latent space is defined as a set of
vectors of given dimensionality. Such representations are not easily
interpretable and do not capture spatial information of image content directly.
In this work, we define a family of spatial latent spaces for StyleGAN2,
capable of capturing more details and representing images that are
out-of-sample in terms of the number and arrangement of object parts, such as
an image of multiple faces or a face with more than two eyes. We propose a
method for encoding images into our spaces, together with an attribute model
capable of performing attribute editing in these spaces. We show that our
spaces are effective for image manipulation and encode semantic information
well. Our approach can be used on pre-trained generator models, and attribute
edition can be done using pre-generated direction vectors making the barrier to
entry for experimentation and use extremely low. We propose a regularization
method for optimizing latent representations, which equalizes distributions of
parts of latent spaces, making representations much closer to generated ones.
We use it for encoding images into spatial spaces to obtain significant
improvement in quality while keeping semantics and ability to use our attribute
model for edition purposes. In total, using our methods gives encoding quality
boost even as high as 30% in terms of LPIPS score comparing to standard
methods, while keeping semantics. Additionally, we propose a StyleGAN2 training
procedure on our spatial latent spaces, together with a custom spatial latent
representation distribution to make spatially closer elements in the
representation more dependent on each other than farther elements. Such
approach improves the FID score by 29% on SpaceNet, and is able to generate
consistent images of arbitrary sizes on spatially homogeneous datasets, like
satellite imagery.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 20:01:11 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Sypetkowski",
"Maciej",
""
]
] | TITLE: Spatial Latent Representations in Generative Adversarial Networks for
Image Generation
ABSTRACT: In the majority of GAN architectures, the latent space is defined as a set of
vectors of given dimensionality. Such representations are not easily
interpretable and do not capture spatial information of image content directly.
In this work, we define a family of spatial latent spaces for StyleGAN2,
capable of capturing more details and representing images that are
out-of-sample in terms of the number and arrangement of object parts, such as
an image of multiple faces or a face with more than two eyes. We propose a
method for encoding images into our spaces, together with an attribute model
capable of performing attribute editing in these spaces. We show that our
spaces are effective for image manipulation and encode semantic information
well. Our approach can be used on pre-trained generator models, and attribute
edition can be done using pre-generated direction vectors making the barrier to
entry for experimentation and use extremely low. We propose a regularization
method for optimizing latent representations, which equalizes distributions of
parts of latent spaces, making representations much closer to generated ones.
We use it for encoding images into spatial spaces to obtain significant
improvement in quality while keeping semantics and ability to use our attribute
model for edition purposes. In total, using our methods gives encoding quality
boost even as high as 30% in terms of LPIPS score comparing to standard
methods, while keeping semantics. Additionally, we propose a StyleGAN2 training
procedure on our spatial latent spaces, together with a custom spatial latent
representation distribution to make spatially closer elements in the
representation more dependent on each other than farther elements. Such
approach improves the FID score by 29% on SpaceNet, and is able to generate
consistent images of arbitrary sizes on spatially homogeneous datasets, like
satellite imagery.
| no_new_dataset | 0.953144 |
2303.14566 | Hanlin Mo | Hanlin Mo, Hongxiang Hao, Guoying Zhao | Image Moment Invariants to Rotational Motion Blur | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Rotational motion blur caused by the circular motion of the camera or/and
object is common in life. Identifying objects from images affected by
rotational motion blur is challenging because this image degradation severely
impacts image quality. Therefore, it is meaningful to develop image invariant
features under rotational motion blur and then use them in practical tasks,
such as object classification and template matching. This paper proposes a
novel method to generate image moment invariants under general rotational
motion blur and provides some instances. Further, we achieve their invariance
to similarity transform. To the best of our knowledge, this is the first time
that moment invariants for rotational motion blur have been proposed in the
literature. We conduct extensive experiments on various image datasets
disturbed by similarity transform and rotational motion blur to test these
invariants' numerical stability and robustness to image noise. We also
demonstrate their performance in image classification and handwritten digit
recognition. Current state-of-the-art blur moment invariants and deep neural
networks are chosen for comparison. Our results show that the moment invariants
proposed in this paper significantly outperform other features in various
tasks.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 21:23:42 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Mo",
"Hanlin",
""
],
[
"Hao",
"Hongxiang",
""
],
[
"Zhao",
"Guoying",
""
]
] | TITLE: Image Moment Invariants to Rotational Motion Blur
ABSTRACT: Rotational motion blur caused by the circular motion of the camera or/and
object is common in life. Identifying objects from images affected by
rotational motion blur is challenging because this image degradation severely
impacts image quality. Therefore, it is meaningful to develop image invariant
features under rotational motion blur and then use them in practical tasks,
such as object classification and template matching. This paper proposes a
novel method to generate image moment invariants under general rotational
motion blur and provides some instances. Further, we achieve their invariance
to similarity transform. To the best of our knowledge, this is the first time
that moment invariants for rotational motion blur have been proposed in the
literature. We conduct extensive experiments on various image datasets
disturbed by similarity transform and rotational motion blur to test these
invariants' numerical stability and robustness to image noise. We also
demonstrate their performance in image classification and handwritten digit
recognition. Current state-of-the-art blur moment invariants and deep neural
networks are chosen for comparison. Our results show that the moment invariants
proposed in this paper significantly outperform other features in various
tasks.
| no_new_dataset | 0.951774 |
2303.14569 | Artsiom Sanakoyeu | Albert Pumarola, Artsiom Sanakoyeu, Lior Yariv, Ali Thabet, Yaron
Lipman | VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids | Published in NeurIPS 2022 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Surface reconstruction has been seeing a lot of progress lately by utilizing
Implicit Neural Representations (INRs). Despite their success, INRs often
introduce hard to control inductive bias (i.e., the solution surface can
exhibit unexplainable behaviours), have costly inference, and are slow to
train. The goal of this work is to show that replacing neural networks with
simple grid functions, along with two novel geometric priors achieve comparable
results to INRs, with instant inference, and improved training times. To that
end we introduce VisCo Grids: a grid-based surface reconstruction method
incorporating Viscosity and Coarea priors. Intuitively, the Viscosity prior
replaces the smoothness inductive bias of INRs, while the Coarea favors a
minimal area solution. Experimenting with VisCo Grids on a standard
reconstruction baseline provided comparable results to the best performing INRs
on this dataset.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 21:32:17 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Pumarola",
"Albert",
""
],
[
"Sanakoyeu",
"Artsiom",
""
],
[
"Yariv",
"Lior",
""
],
[
"Thabet",
"Ali",
""
],
[
"Lipman",
"Yaron",
""
]
] | TITLE: VisCo Grids: Surface Reconstruction with Viscosity and Coarea Grids
ABSTRACT: Surface reconstruction has been seeing a lot of progress lately by utilizing
Implicit Neural Representations (INRs). Despite their success, INRs often
introduce hard to control inductive bias (i.e., the solution surface can
exhibit unexplainable behaviours), have costly inference, and are slow to
train. The goal of this work is to show that replacing neural networks with
simple grid functions, along with two novel geometric priors achieve comparable
results to INRs, with instant inference, and improved training times. To that
end we introduce VisCo Grids: a grid-based surface reconstruction method
incorporating Viscosity and Coarea priors. Intuitively, the Viscosity prior
replaces the smoothness inductive bias of INRs, while the Coarea favors a
minimal area solution. Experimenting with VisCo Grids on a standard
reconstruction baseline provided comparable results to the best performing INRs
on this dataset.
| no_new_dataset | 0.946794 |
2303.14587 | Shuhong Chen | Shuhong Chen, Kevin Zhang, Yichun Shi, Heng Wang, Yiheng Zhu, Guoxian
Song, Sizhe An, Janus Kristjansson, Xiao Yang, Matthias Zwicker | PAniC-3D: Stylized Single-view 3D Reconstruction from Portraits of Anime
Characters | CVPR 2023, code release:
https://github.com/ShuhongChen/panic3d-anime-reconstruction | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We propose PAniC-3D, a system to reconstruct stylized 3D character heads
directly from illustrated (p)ortraits of (ani)me (c)haracters. Our anime-style
domain poses unique challenges to single-view reconstruction; compared to
natural images of human heads, character portrait illustrations have hair and
accessories with more complex and diverse geometry, and are shaded with
non-photorealistic contour lines. In addition, there is a lack of both 3D model
and portrait illustration data suitable to train and evaluate this ambiguous
stylized reconstruction task. Facing these challenges, our proposed PAniC-3D
architecture crosses the illustration-to-3D domain gap with a line-filling
model, and represents sophisticated geometries with a volumetric radiance
field. We train our system with two large new datasets (11.2k Vroid 3D models,
1k Vtuber portrait illustrations), and evaluate on a novel AnimeRecon benchmark
of illustration-to-3D pairs. PAniC-3D significantly outperforms baseline
methods, and provides data to establish the task of stylized reconstruction
from portrait illustrations.
| [
{
"version": "v1",
"created": "Sat, 25 Mar 2023 23:36:17 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Shuhong",
""
],
[
"Zhang",
"Kevin",
""
],
[
"Shi",
"Yichun",
""
],
[
"Wang",
"Heng",
""
],
[
"Zhu",
"Yiheng",
""
],
[
"Song",
"Guoxian",
""
],
[
"An",
"Sizhe",
""
],
[
"Kristjansson",
"Janus",
""
],
[
"Yang",
"Xiao",
""
],
[
"Zwicker",
"Matthias",
""
]
] | TITLE: PAniC-3D: Stylized Single-view 3D Reconstruction from Portraits of Anime
Characters
ABSTRACT: We propose PAniC-3D, a system to reconstruct stylized 3D character heads
directly from illustrated (p)ortraits of (ani)me (c)haracters. Our anime-style
domain poses unique challenges to single-view reconstruction; compared to
natural images of human heads, character portrait illustrations have hair and
accessories with more complex and diverse geometry, and are shaded with
non-photorealistic contour lines. In addition, there is a lack of both 3D model
and portrait illustration data suitable to train and evaluate this ambiguous
stylized reconstruction task. Facing these challenges, our proposed PAniC-3D
architecture crosses the illustration-to-3D domain gap with a line-filling
model, and represents sophisticated geometries with a volumetric radiance
field. We train our system with two large new datasets (11.2k Vroid 3D models,
1k Vtuber portrait illustrations), and evaluate on a novel AnimeRecon benchmark
of illustration-to-3D pairs. PAniC-3D significantly outperforms baseline
methods, and provides data to establish the task of stylized reconstruction
from portrait illustrations.
| new_dataset | 0.953449 |
2303.14589 | Bradford Windsor | Brad Windsor, Luke Martin, Anand Tyagi | SASS: Data and Methods for Subject Aware Sentence Simplification | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Sentence simplification tends to focus on the generic simplification of
sentences by making them more readable and easier to understand. This paper
provides a dataset aimed at training models that perform subject aware sentence
simplifications rather than simplifying sentences as a whole. We also test
models on that dataset which are inspired by model architecture used in
abstractive summarization. We hand generated portions of the data and augment
the dataset by further manipulating those hand written simplifications. Our
results show that data-augmentation, data-masking, and model architecture
choices used in summarization provide a solid baseline for comparison on
subject aware simplification.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 00:02:25 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Windsor",
"Brad",
""
],
[
"Martin",
"Luke",
""
],
[
"Tyagi",
"Anand",
""
]
] | TITLE: SASS: Data and Methods for Subject Aware Sentence Simplification
ABSTRACT: Sentence simplification tends to focus on the generic simplification of
sentences by making them more readable and easier to understand. This paper
provides a dataset aimed at training models that perform subject aware sentence
simplifications rather than simplifying sentences as a whole. We also test
models on that dataset which are inspired by model architecture used in
abstractive summarization. We hand generated portions of the data and augment
the dataset by further manipulating those hand written simplifications. Our
results show that data-augmentation, data-masking, and model architecture
choices used in summarization provide a solid baseline for comparison on
subject aware simplification.
| new_dataset | 0.957991 |
2303.14593 | Hao Shi | Hao Shi, Masato Mimura, Longbiao Wang, Jianwu Dang, Tatsuya Kawahara | Time-domain Speech Enhancement Assisted by Multi-resolution Frequency
Encoder and Decoder | null | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Time-domain speech enhancement (SE) has recently been intensively
investigated. Among recent works, DEMUCS introduces multi-resolution STFT loss
to enhance performance. However, some resolutions used for STFT contain
non-stationary signals, and it is challenging to learn multi-resolution
frequency losses simultaneously with only one output. For better use of
multi-resolution frequency information, we supplement multiple spectrograms in
different frame lengths into the time-domain encoders. They extract stationary
frequency information in both narrowband and wideband. We also adopt multiple
decoder outputs, each of which computes its corresponding resolution frequency
loss. Experimental results show that (1) it is more effective to fuse
stationary frequency features than non-stationary features in the encoder, and
(2) the multiple outputs consistent with the frequency loss improve
performance. Experiments on the Voice-Bank dataset show that the proposed
method obtained a 0.14 PESQ improvement.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 00:30:06 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Shi",
"Hao",
""
],
[
"Mimura",
"Masato",
""
],
[
"Wang",
"Longbiao",
""
],
[
"Dang",
"Jianwu",
""
],
[
"Kawahara",
"Tatsuya",
""
]
] | TITLE: Time-domain Speech Enhancement Assisted by Multi-resolution Frequency
Encoder and Decoder
ABSTRACT: Time-domain speech enhancement (SE) has recently been intensively
investigated. Among recent works, DEMUCS introduces multi-resolution STFT loss
to enhance performance. However, some resolutions used for STFT contain
non-stationary signals, and it is challenging to learn multi-resolution
frequency losses simultaneously with only one output. For better use of
multi-resolution frequency information, we supplement multiple spectrograms in
different frame lengths into the time-domain encoders. They extract stationary
frequency information in both narrowband and wideband. We also adopt multiple
decoder outputs, each of which computes its corresponding resolution frequency
loss. Experimental results show that (1) it is more effective to fuse
stationary frequency features than non-stationary features in the encoder, and
(2) the multiple outputs consistent with the frequency loss improve
performance. Experiments on the Voice-Bank dataset show that the proposed
method obtained a 0.14 PESQ improvement.
| no_new_dataset | 0.947039 |
2303.14601 | Yupei Liu | Jinyuan Jia and Yupei Liu and Yuepeng Hu and Neil Zhenqiang Gong | PORE: Provably Robust Recommender Systems against Data Poisoning Attacks | To appear in USENIX Security Symposium, 2023 | null | null | null | cs.CR cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Data poisoning attacks spoof a recommender system to make arbitrary,
attacker-desired recommendations via injecting fake users with carefully
crafted rating scores into the recommender system. We envision a cat-and-mouse
game for such data poisoning attacks and their defenses, i.e., new defenses are
designed to defend against existing attacks and new attacks are designed to
break them. To prevent such a cat-and-mouse game, we propose PORE, the first
framework to build provably robust recommender systems in this work. PORE can
transform any existing recommender system to be provably robust against any
untargeted data poisoning attacks, which aim to reduce the overall performance
of a recommender system. Suppose PORE recommends top-$N$ items to a user when
there is no attack. We prove that PORE still recommends at least $r$ of the $N$
items to the user under any data poisoning attack, where $r$ is a function of
the number of fake users in the attack. Moreover, we design an efficient
algorithm to compute $r$ for each user. We empirically evaluate PORE on popular
benchmark datasets.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 01:38:11 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jia",
"Jinyuan",
""
],
[
"Liu",
"Yupei",
""
],
[
"Hu",
"Yuepeng",
""
],
[
"Gong",
"Neil Zhenqiang",
""
]
] | TITLE: PORE: Provably Robust Recommender Systems against Data Poisoning Attacks
ABSTRACT: Data poisoning attacks spoof a recommender system to make arbitrary,
attacker-desired recommendations via injecting fake users with carefully
crafted rating scores into the recommender system. We envision a cat-and-mouse
game for such data poisoning attacks and their defenses, i.e., new defenses are
designed to defend against existing attacks and new attacks are designed to
break them. To prevent such a cat-and-mouse game, we propose PORE, the first
framework to build provably robust recommender systems in this work. PORE can
transform any existing recommender system to be provably robust against any
untargeted data poisoning attacks, which aim to reduce the overall performance
of a recommender system. Suppose PORE recommends top-$N$ items to a user when
there is no attack. We prove that PORE still recommends at least $r$ of the $N$
items to the user under any data poisoning attack, where $r$ is a function of
the number of fake users in the attack. Moreover, we design an efficient
algorithm to compute $r$ for each user. We empirically evaluate PORE on popular
benchmark datasets.
| no_new_dataset | 0.943867 |
2303.14617 | Hongyu Ren | Hongyu Ren, Mikhail Galkin, Michael Cochez, Zhaocheng Zhu, Jure
Leskovec | Neural Graph Reasoning: Complex Logical Query Answering Meets Graph
Databases | null | null | null | null | cs.DB cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Complex logical query answering (CLQA) is a recently emerged task of graph
machine learning that goes beyond simple one-hop link prediction and solves a
far more complex task of multi-hop logical reasoning over massive, potentially
incomplete graphs in a latent space. The task received a significant traction
in the community; numerous works expanded the field along theoretical and
practical axes to tackle different types of complex queries and graph
modalities with efficient systems. In this paper, we provide a holistic survey
of CLQA with a detailed taxonomy studying the field from multiple angles,
including graph types (modality, reasoning domain, background semantics),
modeling aspects (encoder, processor, decoder), supported queries (operators,
patterns, projected variables), datasets, evaluation metrics, and applications.
Refining the CLQA task, we introduce the concept of Neural Graph Databases
(NGDBs). Extending the idea of graph databases (graph DBs), NGDB consists of a
Neural Graph Storage and a Neural Graph Engine. Inside Neural Graph Storage, we
design a graph store, a feature store, and further embed information in a
latent embedding store using an encoder. Given a query, Neural Query Engine
learns how to perform query planning and execution in order to efficiently
retrieve the correct results by interacting with the Neural Graph Storage.
Compared with traditional graph DBs, NGDBs allow for a flexible and unified
modeling of features in diverse modalities using the embedding store. Moreover,
when the graph is incomplete, they can provide robust retrieval of answers
which a normal graph DB cannot recover. Finally, we point out promising
directions, unsolved problems and applications of NGDB for future research.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 04:03:37 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ren",
"Hongyu",
""
],
[
"Galkin",
"Mikhail",
""
],
[
"Cochez",
"Michael",
""
],
[
"Zhu",
"Zhaocheng",
""
],
[
"Leskovec",
"Jure",
""
]
] | TITLE: Neural Graph Reasoning: Complex Logical Query Answering Meets Graph
Databases
ABSTRACT: Complex logical query answering (CLQA) is a recently emerged task of graph
machine learning that goes beyond simple one-hop link prediction and solves a
far more complex task of multi-hop logical reasoning over massive, potentially
incomplete graphs in a latent space. The task received a significant traction
in the community; numerous works expanded the field along theoretical and
practical axes to tackle different types of complex queries and graph
modalities with efficient systems. In this paper, we provide a holistic survey
of CLQA with a detailed taxonomy studying the field from multiple angles,
including graph types (modality, reasoning domain, background semantics),
modeling aspects (encoder, processor, decoder), supported queries (operators,
patterns, projected variables), datasets, evaluation metrics, and applications.
Refining the CLQA task, we introduce the concept of Neural Graph Databases
(NGDBs). Extending the idea of graph databases (graph DBs), NGDB consists of a
Neural Graph Storage and a Neural Graph Engine. Inside Neural Graph Storage, we
design a graph store, a feature store, and further embed information in a
latent embedding store using an encoder. Given a query, Neural Query Engine
learns how to perform query planning and execution in order to efficiently
retrieve the correct results by interacting with the Neural Graph Storage.
Compared with traditional graph DBs, NGDBs allow for a flexible and unified
modeling of features in diverse modalities using the embedding store. Moreover,
when the graph is incomplete, they can provide robust retrieval of answers
which a normal graph DB cannot recover. Finally, we point out promising
directions, unsolved problems and applications of NGDB for future research.
| no_new_dataset | 0.946695 |
2303.14626 | Yukang Zhang | Yukang Zhang, Yan Yan, Jie Li, Hanzi Wang | MRCN: A Novel Modality Restitution and Compensation Network for
Visible-Infrared Person Re-identification | Accepted by AAAI-2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visible-infrared person re-identification (VI-ReID), which aims to search
identities across different spectra, is a challenging task due to large
cross-modality discrepancy between visible and infrared images. The key to
reduce the discrepancy is to filter out identity-irrelevant interference and
effectively learn modality-invariant person representations. In this paper, we
propose a novel Modality Restitution and Compensation Network (MRCN) to narrow
the gap between the two modalities. Specifically, we first reduce the modality
discrepancy by using two Instance Normalization (IN) layers. Next, to reduce
the influence of IN layers on removing discriminative information and to reduce
modality differences, we propose a Modality Restitution Module (MRM) and a
Modality Compensation Module (MCM) to respectively distill modality-irrelevant
and modality-relevant features from the removed information. Then, the
modality-irrelevant features are used to restitute to the normalized visible
and infrared features, while the modality-relevant features are used to
compensate for the features of the other modality. Furthermore, to better
disentangle the modality-relevant features and the modality-irrelevant
features, we propose a novel Center-Quadruplet Causal (CQC) loss to encourage
the network to effectively learn the modality-relevant features and the
modality-irrelevant features. Extensive experiments are conducted to validate
the superiority of our method on the challenging SYSU-MM01 and RegDB datasets.
More remarkably, our method achieves 95.1% in terms of Rank-1 and 89.2% in
terms of mAP on the RegDB dataset.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 05:03:18 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Yukang",
""
],
[
"Yan",
"Yan",
""
],
[
"Li",
"Jie",
""
],
[
"Wang",
"Hanzi",
""
]
] | TITLE: MRCN: A Novel Modality Restitution and Compensation Network for
Visible-Infrared Person Re-identification
ABSTRACT: Visible-infrared person re-identification (VI-ReID), which aims to search
identities across different spectra, is a challenging task due to large
cross-modality discrepancy between visible and infrared images. The key to
reduce the discrepancy is to filter out identity-irrelevant interference and
effectively learn modality-invariant person representations. In this paper, we
propose a novel Modality Restitution and Compensation Network (MRCN) to narrow
the gap between the two modalities. Specifically, we first reduce the modality
discrepancy by using two Instance Normalization (IN) layers. Next, to reduce
the influence of IN layers on removing discriminative information and to reduce
modality differences, we propose a Modality Restitution Module (MRM) and a
Modality Compensation Module (MCM) to respectively distill modality-irrelevant
and modality-relevant features from the removed information. Then, the
modality-irrelevant features are used to restitute to the normalized visible
and infrared features, while the modality-relevant features are used to
compensate for the features of the other modality. Furthermore, to better
disentangle the modality-relevant features and the modality-irrelevant
features, we propose a novel Center-Quadruplet Causal (CQC) loss to encourage
the network to effectively learn the modality-relevant features and the
modality-irrelevant features. Extensive experiments are conducted to validate
the superiority of our method on the challenging SYSU-MM01 and RegDB datasets.
More remarkably, our method achieves 95.1% in terms of Rank-1 and 89.2% in
terms of mAP on the RegDB dataset.
| no_new_dataset | 0.951504 |
2303.14644 | Joya Chen | Joya Chen, Difei Gao, Kevin Qinghong Lin, Mike Zheng Shou | Affordance Grounding from Demonstration Video to Target Image | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Humans excel at learning from expert demonstrations and solving their own
problems. To equip intelligent robots and assistants, such as AR glasses, with
this ability, it is essential to ground human hand interactions (i.e.,
affordances) from demonstration videos and apply them to a target image like a
user's AR glass view. The video-to-image affordance grounding task is
challenging due to (1) the need to predict fine-grained affordances, and (2)
the limited training data, which inadequately covers video-image discrepancies
and negatively impacts grounding. To tackle them, we propose Affordance
Transformer (Afformer), which has a fine-grained transformer-based decoder that
gradually refines affordance grounding. Moreover, we introduce Mask Affordance
Hand (MaskAHand), a self-supervised pre-training technique for synthesizing
video-image data and simulating context changes, enhancing affordance grounding
across video-image discrepancies. Afformer with MaskAHand pre-training achieves
state-of-the-art performance on multiple benchmarks, including a substantial
37% improvement on the OPRA dataset. Code is made available at
https://github.com/showlab/afformer.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 07:02:41 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Joya",
""
],
[
"Gao",
"Difei",
""
],
[
"Lin",
"Kevin Qinghong",
""
],
[
"Shou",
"Mike Zheng",
""
]
] | TITLE: Affordance Grounding from Demonstration Video to Target Image
ABSTRACT: Humans excel at learning from expert demonstrations and solving their own
problems. To equip intelligent robots and assistants, such as AR glasses, with
this ability, it is essential to ground human hand interactions (i.e.,
affordances) from demonstration videos and apply them to a target image like a
user's AR glass view. The video-to-image affordance grounding task is
challenging due to (1) the need to predict fine-grained affordances, and (2)
the limited training data, which inadequately covers video-image discrepancies
and negatively impacts grounding. To tackle them, we propose Affordance
Transformer (Afformer), which has a fine-grained transformer-based decoder that
gradually refines affordance grounding. Moreover, we introduce Mask Affordance
Hand (MaskAHand), a self-supervised pre-training technique for synthesizing
video-image data and simulating context changes, enhancing affordance grounding
across video-image discrepancies. Afformer with MaskAHand pre-training achieves
state-of-the-art performance on multiple benchmarks, including a substantial
37% improvement on the OPRA dataset. Code is made available at
https://github.com/showlab/afformer.
| no_new_dataset | 0.949669 |
2303.14645 | Dianyi Yang | Dianyi Yang, Jiadong Tang, Yu Gao, Yi Yang, Mengyin Fu | Sector Patch Embedding: An Embedding Module Conforming to The Distortion
Pattern of Fisheye Image | null | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fisheye cameras suffer from image distortion while having a large field of
view(LFOV). And this fact leads to poor performance on some fisheye vision
tasks. One of the solutions is to optimize the current vision algorithm for
fisheye images. However, most of the CNN-based methods and the
Transformer-based methods lack the capability of leveraging distortion
information efficiently. In this work, we propose a novel patch embedding
method called Sector Patch Embedding(SPE), conforming to the distortion pattern
of the fisheye image. Furthermore, we put forward a synthetic fisheye dataset
based on the ImageNet-1K and explore the performance of several Transformer
models on the dataset. The classification top-1 accuracy of ViT and PVT is
improved by 0.75% and 2.8% with SPE respectively. The experiments show that the
proposed sector patch embedding method can better perceive distortion and
extract features on the fisheye images. Our method can be easily adopted to
other Transformer-based models. Source code is at
https://github.com/IN2-ViAUn/Sector-Patch-Embedding.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 07:20:02 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yang",
"Dianyi",
""
],
[
"Tang",
"Jiadong",
""
],
[
"Gao",
"Yu",
""
],
[
"Yang",
"Yi",
""
],
[
"Fu",
"Mengyin",
""
]
] | TITLE: Sector Patch Embedding: An Embedding Module Conforming to The Distortion
Pattern of Fisheye Image
ABSTRACT: Fisheye cameras suffer from image distortion while having a large field of
view(LFOV). And this fact leads to poor performance on some fisheye vision
tasks. One of the solutions is to optimize the current vision algorithm for
fisheye images. However, most of the CNN-based methods and the
Transformer-based methods lack the capability of leveraging distortion
information efficiently. In this work, we propose a novel patch embedding
method called Sector Patch Embedding(SPE), conforming to the distortion pattern
of the fisheye image. Furthermore, we put forward a synthetic fisheye dataset
based on the ImageNet-1K and explore the performance of several Transformer
models on the dataset. The classification top-1 accuracy of ViT and PVT is
improved by 0.75% and 2.8% with SPE respectively. The experiments show that the
proposed sector patch embedding method can better perceive distortion and
extract features on the fisheye images. Our method can be easily adopted to
other Transformer-based models. Source code is at
https://github.com/IN2-ViAUn/Sector-Patch-Embedding.
| no_new_dataset | 0.945851 |
2303.14646 | Dacheng Wen | Dacheng Wen, Yupeng Li, Francis C.M. Lau | A Survey of Machine Learning-Based Ride-Hailing Planning | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Ride-hailing is a sustainable transportation paradigm where riders access
door-to-door traveling services through a mobile phone application, which has
attracted a colossal amount of usage. There are two major planning tasks in a
ride-hailing system: (1) matching, i.e., assigning available vehicles to pick
up the riders, and (2) repositioning, i.e., proactively relocating vehicles to
certain locations to balance the supply and demand of ride-hailing services.
Recently, many studies of ride-hailing planning that leverage machine learning
techniques have emerged. In this article, we present a comprehensive overview
on latest developments of machine learning-based ride-hailing planning. To
offer a clear and structured review, we introduce a taxonomy into which we
carefully fit the different categories of related works according to the types
of their planning tasks and solution schemes, which include collective
matching, distributed matching, collective repositioning, distributed
repositioning, and joint matching and repositioning. We further shed light on
many real-world datasets and simulators that are indispensable for empirical
studies on machine learning-based ride-hailing planning strategies. At last, we
propose several promising research directions for this rapidly growing research
and practical field.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 07:21:03 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wen",
"Dacheng",
""
],
[
"Li",
"Yupeng",
""
],
[
"Lau",
"Francis C. M.",
""
]
] | TITLE: A Survey of Machine Learning-Based Ride-Hailing Planning
ABSTRACT: Ride-hailing is a sustainable transportation paradigm where riders access
door-to-door traveling services through a mobile phone application, which has
attracted a colossal amount of usage. There are two major planning tasks in a
ride-hailing system: (1) matching, i.e., assigning available vehicles to pick
up the riders, and (2) repositioning, i.e., proactively relocating vehicles to
certain locations to balance the supply and demand of ride-hailing services.
Recently, many studies of ride-hailing planning that leverage machine learning
techniques have emerged. In this article, we present a comprehensive overview
on latest developments of machine learning-based ride-hailing planning. To
offer a clear and structured review, we introduce a taxonomy into which we
carefully fit the different categories of related works according to the types
of their planning tasks and solution schemes, which include collective
matching, distributed matching, collective repositioning, distributed
repositioning, and joint matching and repositioning. We further shed light on
many real-world datasets and simulators that are indispensable for empirical
studies on machine learning-based ride-hailing planning strategies. At last, we
propose several promising research directions for this rapidly growing research
and practical field.
| no_new_dataset | 0.947381 |
2303.14647 | Behrouz Minaei-Bidgoli | Najmeh Torabian, Behrouz Minaei-Bidgoli and Mohsen Jahanshahi | Farspredict: A benchmark dataset for link prediction | 13 pages, 3 figures, 1 algorithm and 5 tables | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Link prediction with knowledge graph embedding (KGE) is a popular method for
knowledge graph completion. Furthermore, training KGEs on non-English knowledge
graph promote knowledge extraction and knowledge graph reasoning in the context
of these languages. However, many challenges in non-English KGEs pose to
learning a low-dimensional representation of a knowledge graph's entities and
relations. This paper proposes "Farspredict" a Persian knowledge graph based on
Farsbase (the most comprehensive knowledge graph in Persian). It also explains
how the knowledge graph structure affects link prediction accuracy in KGE. To
evaluate Farspredict, we implemented the popular models of KGE on it and
compared the results with Freebase. Given the analysis results, some
optimizations on the knowledge graph are carried out to improve its
functionality in the KGE. As a result, a new Persian knowledge graph is
achieved. Implementation results in the KGE models on Farspredict outperforming
Freebases in many cases. At last, we discuss what improvements could be
effective in enhancing the quality of Farspredict and how much it improves.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 07:41:26 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Torabian",
"Najmeh",
""
],
[
"Minaei-Bidgoli",
"Behrouz",
""
],
[
"Jahanshahi",
"Mohsen",
""
]
] | TITLE: Farspredict: A benchmark dataset for link prediction
ABSTRACT: Link prediction with knowledge graph embedding (KGE) is a popular method for
knowledge graph completion. Furthermore, training KGEs on non-English knowledge
graph promote knowledge extraction and knowledge graph reasoning in the context
of these languages. However, many challenges in non-English KGEs pose to
learning a low-dimensional representation of a knowledge graph's entities and
relations. This paper proposes "Farspredict" a Persian knowledge graph based on
Farsbase (the most comprehensive knowledge graph in Persian). It also explains
how the knowledge graph structure affects link prediction accuracy in KGE. To
evaluate Farspredict, we implemented the popular models of KGE on it and
compared the results with Freebase. Given the analysis results, some
optimizations on the knowledge graph are carried out to improve its
functionality in the KGE. As a result, a new Persian knowledge graph is
achieved. Implementation results in the KGE models on Farspredict outperforming
Freebases in many cases. At last, we discuss what improvements could be
effective in enhancing the quality of Farspredict and how much it improves.
| new_dataset | 0.964018 |
2303.14652 | Bohao Peng | Bohao Peng, Zhuotao Tian, Xiaoyang Wu, Chenyao Wang, Shu Liu, Jingyong
Su, Jiaya Jia | Hierarchical Dense Correlation Distillation for Few-Shot Segmentation | to be published in CVPR 2023, code is available at
\url{https://github.com/Pbihao/HDMNet} | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Few-shot semantic segmentation (FSS) aims to form class-agnostic models
segmenting unseen classes with only a handful of annotations. Previous methods
limited to the semantic feature and prototype representation suffer from coarse
segmentation granularity and train-set overfitting. In this work, we design
Hierarchically Decoupled Matching Network (HDMNet) mining pixel-level support
correlation based on the transformer architecture. The self-attention modules
are used to assist in establishing hierarchical dense features, as a means to
accomplish the cascade matching between query and support features. Moreover,
we propose a matching module to reduce train-set overfitting and introduce
correlation distillation leveraging semantic correspondence from coarse
resolution to boost fine-grained segmentation. Our method performs decently in
experiments. We achieve $50.0\%$ mIoU on \coco~dataset one-shot setting and
$56.0\%$ on five-shot segmentation, respectively.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 08:13:12 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Peng",
"Bohao",
""
],
[
"Tian",
"Zhuotao",
""
],
[
"Wu",
"Xiaoyang",
""
],
[
"Wang",
"Chenyao",
""
],
[
"Liu",
"Shu",
""
],
[
"Su",
"Jingyong",
""
],
[
"Jia",
"Jiaya",
""
]
] | TITLE: Hierarchical Dense Correlation Distillation for Few-Shot Segmentation
ABSTRACT: Few-shot semantic segmentation (FSS) aims to form class-agnostic models
segmenting unseen classes with only a handful of annotations. Previous methods
limited to the semantic feature and prototype representation suffer from coarse
segmentation granularity and train-set overfitting. In this work, we design
Hierarchically Decoupled Matching Network (HDMNet) mining pixel-level support
correlation based on the transformer architecture. The self-attention modules
are used to assist in establishing hierarchical dense features, as a means to
accomplish the cascade matching between query and support features. Moreover,
we propose a matching module to reduce train-set overfitting and introduce
correlation distillation leveraging semantic correspondence from coarse
resolution to boost fine-grained segmentation. Our method performs decently in
experiments. We achieve $50.0\%$ mIoU on \coco~dataset one-shot setting and
$56.0\%$ on five-shot segmentation, respectively.
| no_new_dataset | 0.951097 |
2303.14653 | Min Wang | Yingda Guan, Zhengyang Feng, Huiying Chang, Kuo Du, Tingting Li, Min
Wang | SDTracker: Synthetic Data Based Multi-Object Tracking | cvpr2022 workshop | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present SDTracker, a method that harnesses the potential of synthetic data
for multi-object tracking of real-world scenes in a domain generalization and
semi-supervised fashion. First, we use the ImageNet dataset as an auxiliary to
randomize the style of synthetic data. With out-of-domain data, we further
enforce pyramid consistency loss across different "stylized" images from the
same sample to learn domain invariant features. Second, we adopt the
pseudo-labeling method to effectively utilize the unlabeled MOT17 training
data. To obtain high-quality pseudo-labels, we apply proximal policy
optimization (PPO2) algorithm to search confidence thresholds for each
sequence. When using the unlabeled MOT17 training set, combined with the
pure-motion tracking strategy upgraded via developed post-processing, we
finally reach 61.4 HOTA.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 08:21:22 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Guan",
"Yingda",
""
],
[
"Feng",
"Zhengyang",
""
],
[
"Chang",
"Huiying",
""
],
[
"Du",
"Kuo",
""
],
[
"Li",
"Tingting",
""
],
[
"Wang",
"Min",
""
]
] | TITLE: SDTracker: Synthetic Data Based Multi-Object Tracking
ABSTRACT: We present SDTracker, a method that harnesses the potential of synthetic data
for multi-object tracking of real-world scenes in a domain generalization and
semi-supervised fashion. First, we use the ImageNet dataset as an auxiliary to
randomize the style of synthetic data. With out-of-domain data, we further
enforce pyramid consistency loss across different "stylized" images from the
same sample to learn domain invariant features. Second, we adopt the
pseudo-labeling method to effectively utilize the unlabeled MOT17 training
data. To obtain high-quality pseudo-labels, we apply proximal policy
optimization (PPO2) algorithm to search confidence thresholds for each
sequence. When using the unlabeled MOT17 training set, combined with the
pure-motion tracking strategy upgraded via developed post-processing, we
finally reach 61.4 HOTA.
| no_new_dataset | 0.950365 |
2303.14665 | Guandong Xu | Tri Dung Duong, Qian Li, Guandong Xu | Achieving Counterfactual Fairness with Imperfect Structural Causal Model | null | null | null | null | cs.LG cs.AI cs.CY | http://creativecommons.org/licenses/by/4.0/ | Counterfactual fairness alleviates the discrimination between the model
prediction toward an individual in the actual world (observational data) and
that in counterfactual world (i.e., what if the individual belongs to other
sensitive groups). The existing studies need to pre-define the structural
causal model that captures the correlations among variables for counterfactual
inference; however, the underlying causal model is usually unknown and
difficult to be validated in real-world scenarios. Moreover, the
misspecification of the causal model potentially leads to poor performance in
model prediction and thus makes unfair decisions. In this research, we propose
a novel minimax game-theoretic model for counterfactual fairness that can
produce accurate results meanwhile achieve a counterfactually fair decision
with the relaxation of strong assumptions of structural causal models. In
addition, we also theoretically prove the error bound of the proposed minimax
model. Empirical experiments on multiple real-world datasets illustrate our
superior performance in both accuracy and fairness. Source code is available at
\url{https://github.com/tridungduong16/counterfactual_fairness_game_theoretic}.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 09:37:29 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Duong",
"Tri Dung",
""
],
[
"Li",
"Qian",
""
],
[
"Xu",
"Guandong",
""
]
] | TITLE: Achieving Counterfactual Fairness with Imperfect Structural Causal Model
ABSTRACT: Counterfactual fairness alleviates the discrimination between the model
prediction toward an individual in the actual world (observational data) and
that in counterfactual world (i.e., what if the individual belongs to other
sensitive groups). The existing studies need to pre-define the structural
causal model that captures the correlations among variables for counterfactual
inference; however, the underlying causal model is usually unknown and
difficult to be validated in real-world scenarios. Moreover, the
misspecification of the causal model potentially leads to poor performance in
model prediction and thus makes unfair decisions. In this research, we propose
a novel minimax game-theoretic model for counterfactual fairness that can
produce accurate results meanwhile achieve a counterfactually fair decision
with the relaxation of strong assumptions of structural causal models. In
addition, we also theoretically prove the error bound of the proposed minimax
model. Empirical experiments on multiple real-world datasets illustrate our
superior performance in both accuracy and fairness. Source code is available at
\url{https://github.com/tridungduong16/counterfactual_fairness_game_theoretic}.
| no_new_dataset | 0.950595 |
2303.14679 | Yongqi An | Yongqi An, Xu Zhao, Tao Yu, Haiyun Guo, Chaoyang Zhao, Ming Tang,
Jinqiao Wang | ZBS: Zero-shot Background Subtraction via Instance-level Background
Modeling and Foreground Selection | Accepted by CVPR 2023. Code is available at
https://github.com/CASIA-IVA-Lab/ZBS | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Background subtraction (BGS) aims to extract all moving objects in the video
frames to obtain binary foreground segmentation masks. Deep learning has been
widely used in this field. Compared with supervised-based BGS methods,
unsupervised methods have better generalization. However, previous unsupervised
deep learning BGS algorithms perform poorly in sophisticated scenarios such as
shadows or night lights, and they cannot detect objects outside the pre-defined
categories. In this work, we propose an unsupervised BGS algorithm based on
zero-shot object detection called Zero-shot Background Subtraction (ZBS). The
proposed method fully utilizes the advantages of zero-shot object detection to
build the open-vocabulary instance-level background model. Based on it, the
foreground can be effectively extracted by comparing the detection results of
new frames with the background model. ZBS performs well for sophisticated
scenarios, and it has rich and extensible categories. Furthermore, our method
can easily generalize to other tasks, such as abandoned object detection in
unseen environments. We experimentally show that ZBS surpasses state-of-the-art
unsupervised BGS methods by 4.70% F-Measure on the CDnet 2014 dataset. The code
is released at https://github.com/CASIA-IVA-Lab/ZBS.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 10:59:03 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"An",
"Yongqi",
""
],
[
"Zhao",
"Xu",
""
],
[
"Yu",
"Tao",
""
],
[
"Guo",
"Haiyun",
""
],
[
"Zhao",
"Chaoyang",
""
],
[
"Tang",
"Ming",
""
],
[
"Wang",
"Jinqiao",
""
]
] | TITLE: ZBS: Zero-shot Background Subtraction via Instance-level Background
Modeling and Foreground Selection
ABSTRACT: Background subtraction (BGS) aims to extract all moving objects in the video
frames to obtain binary foreground segmentation masks. Deep learning has been
widely used in this field. Compared with supervised-based BGS methods,
unsupervised methods have better generalization. However, previous unsupervised
deep learning BGS algorithms perform poorly in sophisticated scenarios such as
shadows or night lights, and they cannot detect objects outside the pre-defined
categories. In this work, we propose an unsupervised BGS algorithm based on
zero-shot object detection called Zero-shot Background Subtraction (ZBS). The
proposed method fully utilizes the advantages of zero-shot object detection to
build the open-vocabulary instance-level background model. Based on it, the
foreground can be effectively extracted by comparing the detection results of
new frames with the background model. ZBS performs well for sophisticated
scenarios, and it has rich and extensible categories. Furthermore, our method
can easily generalize to other tasks, such as abandoned object detection in
unseen environments. We experimentally show that ZBS surpasses state-of-the-art
unsupervised BGS methods by 4.70% F-Measure on the CDnet 2014 dataset. The code
is released at https://github.com/CASIA-IVA-Lab/ZBS.
| no_new_dataset | 0.947381 |
2303.14706 | Qian Wang | Qian Wang, Yiqun Wang, Michael Birsak, Peter Wonka | BlobGAN-3D: A Spatially-Disentangled 3D-Aware Generative Model for
Indoor Scenes | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | 3D-aware image synthesis has attracted increasing interest as it models the
3D nature of our real world. However, performing realistic object-level editing
of the generated images in the multi-object scenario still remains a challenge.
Recently, a 2D GAN termed BlobGAN has demonstrated great multi-object editing
capabilities on real-world indoor scene datasets. In this work, we propose
BlobGAN-3D, which is a 3D-aware improvement of the original 2D BlobGAN. We
enable explicit camera pose control while maintaining the disentanglement for
individual objects in the scene by extending the 2D blobs into 3D blobs. We
keep the object-level editing capabilities of BlobGAN and in addition allow
flexible control over the 3D location of the objects in the scene. We test our
method on real-world indoor datasets and show that our method can achieve
comparable image quality compared to the 2D BlobGAN and other 3D-aware GAN
baselines while being able to enable camera pose control and object-level
editing in the challenging multi-object real-world scenarios.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 12:23:11 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Qian",
""
],
[
"Wang",
"Yiqun",
""
],
[
"Birsak",
"Michael",
""
],
[
"Wonka",
"Peter",
""
]
] | TITLE: BlobGAN-3D: A Spatially-Disentangled 3D-Aware Generative Model for
Indoor Scenes
ABSTRACT: 3D-aware image synthesis has attracted increasing interest as it models the
3D nature of our real world. However, performing realistic object-level editing
of the generated images in the multi-object scenario still remains a challenge.
Recently, a 2D GAN termed BlobGAN has demonstrated great multi-object editing
capabilities on real-world indoor scene datasets. In this work, we propose
BlobGAN-3D, which is a 3D-aware improvement of the original 2D BlobGAN. We
enable explicit camera pose control while maintaining the disentanglement for
individual objects in the scene by extending the 2D blobs into 3D blobs. We
keep the object-level editing capabilities of BlobGAN and in addition allow
flexible control over the 3D location of the objects in the scene. We test our
method on real-world indoor datasets and show that our method can achieve
comparable image quality compared to the 2D BlobGAN and other 3D-aware GAN
baselines while being able to enable camera pose control and object-level
editing in the challenging multi-object real-world scenarios.
| no_new_dataset | 0.950227 |
2303.14708 | Huiru Wang | Huiru Wang, Xiuhong Li, Zenyu Ren, Dan Yang, chunming Ma | Exploring Multimodal Sentiment Analysis via CBAM Attention and
Double-layer BiLSTM Architecture | null | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Because multimodal data contains more modal information, multimodal sentiment
analysis has become a recent research hotspot. However, redundant information
is easily involved in feature fusion after feature extraction, which has a
certain impact on the feature representation after fusion. Therefore, in this
papaer, we propose a new multimodal sentiment analysis model. In our model, we
use BERT + BiLSTM as new feature extractor to capture the long-distance
dependencies in sentences and consider the position information of input
sequences to obtain richer text features. To remove redundant information and
make the network pay more attention to the correlation between image and text
features, CNN and CBAM attention are added after splicing text features and
picture features, to improve the feature representation ability. On the
MVSA-single dataset and HFM dataset, compared with the baseline model, the ACC
of our model is improved by 1.78% and 1.91%, and the F1 value is enhanced by
3.09% and 2.0%, respectively. The experimental results show that our model
achieves a sound effect, similar to the advanced model.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 12:34:01 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Huiru",
""
],
[
"Li",
"Xiuhong",
""
],
[
"Ren",
"Zenyu",
""
],
[
"Yang",
"Dan",
""
],
[
"Ma",
"chunming",
""
]
] | TITLE: Exploring Multimodal Sentiment Analysis via CBAM Attention and
Double-layer BiLSTM Architecture
ABSTRACT: Because multimodal data contains more modal information, multimodal sentiment
analysis has become a recent research hotspot. However, redundant information
is easily involved in feature fusion after feature extraction, which has a
certain impact on the feature representation after fusion. Therefore, in this
papaer, we propose a new multimodal sentiment analysis model. In our model, we
use BERT + BiLSTM as new feature extractor to capture the long-distance
dependencies in sentences and consider the position information of input
sequences to obtain richer text features. To remove redundant information and
make the network pay more attention to the correlation between image and text
features, CNN and CBAM attention are added after splicing text features and
picture features, to improve the feature representation ability. On the
MVSA-single dataset and HFM dataset, compared with the baseline model, the ACC
of our model is improved by 1.78% and 1.91%, and the F1 value is enhanced by
3.09% and 2.0%, respectively. The experimental results show that our model
achieves a sound effect, similar to the advanced model.
| no_new_dataset | 0.949576 |
2303.14711 | Marcel Reimann | Marcel Reimann, Jungeun Won, Hiroyuki Takahashi, Antonio Yaghy,
Yunchan Hwang, Stefan Ploner, Junhong Lin, Jessica Girgis, Kenneth Lam, Siyu
Chen, Nadia K. Waheed, Andreas Maier, James G. Fujimoto | Unsupervised detection of small hyperreflective features in ultrahigh
resolution optical coherence tomography | Accepted as poster at BVM workshop 2023
(https://www.bvm-workshop.org/). The arXiv version provides full quality
figures. 6 pages content (2 figures) | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recent advances in optical coherence tomography such as the development of
high speed ultrahigh resolution scanners and corresponding signal processing
techniques may reveal new potential biomarkers in retinal diseases. Newly
visible features are, for example, small hyperreflective specks in age-related
macular degeneration. Identifying these new markers is crucial to investigate
potential association with disease progression and treatment outcomes.
Therefore, it is necessary to reliably detect these features in 3D volumetric
scans. Because manual labeling of entire volumes is infeasible a need for
automatic detection arises. Labeled datasets are often not publicly available
and there are usually large variations in scan protocols and scanner types.
Thus, this work focuses on an unsupervised approach that is based on local
peak-detection and random walker segmentation to detect small features on each
B-scan of the volume.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 12:45:42 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Reimann",
"Marcel",
""
],
[
"Won",
"Jungeun",
""
],
[
"Takahashi",
"Hiroyuki",
""
],
[
"Yaghy",
"Antonio",
""
],
[
"Hwang",
"Yunchan",
""
],
[
"Ploner",
"Stefan",
""
],
[
"Lin",
"Junhong",
""
],
[
"Girgis",
"Jessica",
""
],
[
"Lam",
"Kenneth",
""
],
[
"Chen",
"Siyu",
""
],
[
"Waheed",
"Nadia K.",
""
],
[
"Maier",
"Andreas",
""
],
[
"Fujimoto",
"James G.",
""
]
] | TITLE: Unsupervised detection of small hyperreflective features in ultrahigh
resolution optical coherence tomography
ABSTRACT: Recent advances in optical coherence tomography such as the development of
high speed ultrahigh resolution scanners and corresponding signal processing
techniques may reveal new potential biomarkers in retinal diseases. Newly
visible features are, for example, small hyperreflective specks in age-related
macular degeneration. Identifying these new markers is crucial to investigate
potential association with disease progression and treatment outcomes.
Therefore, it is necessary to reliably detect these features in 3D volumetric
scans. Because manual labeling of entire volumes is infeasible a need for
automatic detection arises. Labeled datasets are often not publicly available
and there are usually large variations in scan protocols and scanner types.
Thus, this work focuses on an unsupervised approach that is based on local
peak-detection and random walker segmentation to detect small features on each
B-scan of the volume.
| no_new_dataset | 0.952574 |
2303.14716 | Giovanni Montana | Alex Beeson and Giovanni Montana | Balancing policy constraint and ensemble size in uncertainty-based
offline reinforcement learning | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Offline reinforcement learning agents seek optimal policies from fixed data
sets. With environmental interaction prohibited, agents face significant
challenges in preventing errors in value estimates from compounding and
subsequently causing the learning process to collapse. Uncertainty estimation
using ensembles compensates for this by penalising high-variance value
estimates, allowing agents to learn robust policies based on data-driven
actions. However, the requirement for large ensembles to facilitate sufficient
penalisation results in significant computational overhead. In this work, we
examine the role of policy constraints as a mechanism for regulating
uncertainty, and the corresponding balance between level of constraint and
ensemble size. By incorporating behavioural cloning into policy updates, we
show empirically that sufficient penalisation can be achieved with a much
smaller ensemble size, substantially reducing computational demand while
retaining state-of-the-art performance on benchmarking tasks. Furthermore, we
show how such an approach can facilitate stable online fine tuning, allowing
for continued policy improvement while avoiding severe performance drops.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 13:03:11 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Beeson",
"Alex",
""
],
[
"Montana",
"Giovanni",
""
]
] | TITLE: Balancing policy constraint and ensemble size in uncertainty-based
offline reinforcement learning
ABSTRACT: Offline reinforcement learning agents seek optimal policies from fixed data
sets. With environmental interaction prohibited, agents face significant
challenges in preventing errors in value estimates from compounding and
subsequently causing the learning process to collapse. Uncertainty estimation
using ensembles compensates for this by penalising high-variance value
estimates, allowing agents to learn robust policies based on data-driven
actions. However, the requirement for large ensembles to facilitate sufficient
penalisation results in significant computational overhead. In this work, we
examine the role of policy constraints as a mechanism for regulating
uncertainty, and the corresponding balance between level of constraint and
ensemble size. By incorporating behavioural cloning into policy updates, we
show empirically that sufficient penalisation can be achieved with a much
smaller ensemble size, substantially reducing computational demand while
retaining state-of-the-art performance on benchmarking tasks. Furthermore, we
show how such an approach can facilitate stable online fine tuning, allowing
for continued policy improvement while avoiding severe performance drops.
| no_new_dataset | 0.945901 |
2303.14717 | Jianhui Yu | Jianhui Yu, Hao Zhu, Liming Jiang, Chen Change Loy, Weidong Cai, Wayne
Wu | CelebV-Text: A Large-Scale Facial Text-Video Dataset | Accepted by CVPR2023. Project Page: https://celebv-text.github.io/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Text-driven generation models are flourishing in video generation and
editing. However, face-centric text-to-video generation remains a challenge due
to the lack of a suitable dataset containing high-quality videos and highly
relevant texts. This paper presents CelebV-Text, a large-scale, diverse, and
high-quality dataset of facial text-video pairs, to facilitate research on
facial text-to-video generation tasks. CelebV-Text comprises 70,000 in-the-wild
face video clips with diverse visual content, each paired with 20 texts
generated using the proposed semi-automatic text generation strategy. The
provided texts are of high quality, describing both static and dynamic
attributes precisely. The superiority of CelebV-Text over other datasets is
demonstrated via comprehensive statistical analysis of the videos, texts, and
text-video relevance. The effectiveness and potential of CelebV-Text are
further shown through extensive self-evaluation. A benchmark is constructed
with representative methods to standardize the evaluation of the facial
text-to-video generation task. All data and models are publicly available.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 13:06:35 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yu",
"Jianhui",
""
],
[
"Zhu",
"Hao",
""
],
[
"Jiang",
"Liming",
""
],
[
"Loy",
"Chen Change",
""
],
[
"Cai",
"Weidong",
""
],
[
"Wu",
"Wayne",
""
]
] | TITLE: CelebV-Text: A Large-Scale Facial Text-Video Dataset
ABSTRACT: Text-driven generation models are flourishing in video generation and
editing. However, face-centric text-to-video generation remains a challenge due
to the lack of a suitable dataset containing high-quality videos and highly
relevant texts. This paper presents CelebV-Text, a large-scale, diverse, and
high-quality dataset of facial text-video pairs, to facilitate research on
facial text-to-video generation tasks. CelebV-Text comprises 70,000 in-the-wild
face video clips with diverse visual content, each paired with 20 texts
generated using the proposed semi-automatic text generation strategy. The
provided texts are of high quality, describing both static and dynamic
attributes precisely. The superiority of CelebV-Text over other datasets is
demonstrated via comprehensive statistical analysis of the videos, texts, and
text-video relevance. The effectiveness and potential of CelebV-Text are
further shown through extensive self-evaluation. A benchmark is constructed
with representative methods to standardize the evaluation of the facial
text-to-video generation task. All data and models are publicly available.
| new_dataset | 0.956022 |
2303.14726 | Xiaoming Li | Xiaoming Li, Wangmeng Zuo, Chen Change Loy | Learning Generative Structure Prior for Blind Text Image
Super-resolution | CVPR 2023. Code: https://github.com/csxmli2016/MARCONet | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Blind text image super-resolution (SR) is challenging as one needs to cope
with diverse font styles and unknown degradation. To address the problem,
existing methods perform character recognition in parallel to regularize the SR
task, either through a loss constraint or intermediate feature condition.
Nonetheless, the high-level prior could still fail when encountering severe
degradation. The problem is further compounded given characters of complex
structures, e.g., Chinese characters that combine multiple pictographic or
ideographic symbols into a single character. In this work, we present a novel
prior that focuses more on the character structure. In particular, we learn to
encapsulate rich and diverse structures in a StyleGAN and exploit such
generative structure priors for restoration. To restrict the generative space
of StyleGAN so that it obeys the structure of characters yet remains flexible
in handling different font styles, we store the discrete features for each
character in a codebook. The code subsequently drives the StyleGAN to generate
high-resolution structural details to aid text SR. Compared to priors based on
character recognition, the proposed structure prior exerts stronger
character-specific guidance to restore faithful and precise strokes of a
designated character. Extensive experiments on synthetic and real datasets
demonstrate the compelling performance of the proposed generative structure
prior in facilitating robust text SR.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 13:54:28 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Li",
"Xiaoming",
""
],
[
"Zuo",
"Wangmeng",
""
],
[
"Loy",
"Chen Change",
""
]
] | TITLE: Learning Generative Structure Prior for Blind Text Image
Super-resolution
ABSTRACT: Blind text image super-resolution (SR) is challenging as one needs to cope
with diverse font styles and unknown degradation. To address the problem,
existing methods perform character recognition in parallel to regularize the SR
task, either through a loss constraint or intermediate feature condition.
Nonetheless, the high-level prior could still fail when encountering severe
degradation. The problem is further compounded given characters of complex
structures, e.g., Chinese characters that combine multiple pictographic or
ideographic symbols into a single character. In this work, we present a novel
prior that focuses more on the character structure. In particular, we learn to
encapsulate rich and diverse structures in a StyleGAN and exploit such
generative structure priors for restoration. To restrict the generative space
of StyleGAN so that it obeys the structure of characters yet remains flexible
in handling different font styles, we store the discrete features for each
character in a codebook. The code subsequently drives the StyleGAN to generate
high-resolution structural details to aid text SR. Compared to priors based on
character recognition, the proposed structure prior exerts stronger
character-specific guidance to restore faithful and precise strokes of a
designated character. Extensive experiments on synthetic and real datasets
demonstrate the compelling performance of the proposed generative structure
prior in facilitating robust text SR.
| no_new_dataset | 0.947332 |
2303.14734 | Paolo Bonetti | Paolo Bonetti, Alberto Maria Metelli, Marcello Restelli | Interpretable Linear Dimensionality Reduction based on Bias-Variance
Analysis | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | One of the central issues of several machine learning applications on real
data is the choice of the input features. Ideally, the designer should select
only the relevant, non-redundant features to preserve the complete information
contained in the original dataset, with little collinearity among features and
a smaller dimension. This procedure helps mitigate problems like overfitting
and the curse of dimensionality, which arise when dealing with high-dimensional
problems. On the other hand, it is not desirable to simply discard some
features, since they may still contain information that can be exploited to
improve results. Instead, dimensionality reduction techniques are designed to
limit the number of features in a dataset by projecting them into a
lower-dimensional space, possibly considering all the original features.
However, the projected features resulting from the application of
dimensionality reduction techniques are usually difficult to interpret. In this
paper, we seek to design a principled dimensionality reduction approach that
maintains the interpretability of the resulting features. Specifically, we
propose a bias-variance analysis for linear models and we leverage these
theoretical results to design an algorithm, Linear Correlated Features
Aggregation (LinCFA), which aggregates groups of continuous features with their
average if their correlation is "sufficiently large". In this way, all features
are considered, the dimensionality is reduced and the interpretability is
preserved. Finally, we provide numerical validations of the proposed algorithm
both on synthetic datasets to confirm the theoretical results and on real
datasets to show some promising applications.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 14:30:38 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Bonetti",
"Paolo",
""
],
[
"Metelli",
"Alberto Maria",
""
],
[
"Restelli",
"Marcello",
""
]
] | TITLE: Interpretable Linear Dimensionality Reduction based on Bias-Variance
Analysis
ABSTRACT: One of the central issues of several machine learning applications on real
data is the choice of the input features. Ideally, the designer should select
only the relevant, non-redundant features to preserve the complete information
contained in the original dataset, with little collinearity among features and
a smaller dimension. This procedure helps mitigate problems like overfitting
and the curse of dimensionality, which arise when dealing with high-dimensional
problems. On the other hand, it is not desirable to simply discard some
features, since they may still contain information that can be exploited to
improve results. Instead, dimensionality reduction techniques are designed to
limit the number of features in a dataset by projecting them into a
lower-dimensional space, possibly considering all the original features.
However, the projected features resulting from the application of
dimensionality reduction techniques are usually difficult to interpret. In this
paper, we seek to design a principled dimensionality reduction approach that
maintains the interpretability of the resulting features. Specifically, we
propose a bias-variance analysis for linear models and we leverage these
theoretical results to design an algorithm, Linear Correlated Features
Aggregation (LinCFA), which aggregates groups of continuous features with their
average if their correlation is "sufficiently large". In this way, all features
are considered, the dimensionality is reduced and the interpretability is
preserved. Finally, we provide numerical validations of the proposed algorithm
both on synthetic datasets to confirm the theoretical results and on real
datasets to show some promising applications.
| no_new_dataset | 0.943191 |
2303.14742 | Baochang Ma | Yunjie Ji, Yong Deng, Yan Gong, Yiping Peng, Qiang Niu, Lei Zhang,
Baochang Ma, Xiangang Li | Exploring the Impact of Instruction Data Scaling on Large Language
Models: An Empirical Study on Real-World Use Cases | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The success of ChatGPT has recently attracted numerous efforts to replicate
it, with instruction-tuning strategies being a key factor in achieving
remarkable results. Instruction-tuning not only significantly enhances the
model's performance and generalization but also makes the model's generated
results more consistent with human speech patterns. However current research
rarely studies the impact of different amounts of instruction data on model
performance, especially in the real-world use cases. In this paper we explore
the performance of large language models based on instruction tuning across
different scales of instruction data. An evaluation dataset consisting of 12
major online use cases is constructed in the experiment. With Bloomz-7B1-mt as
the base model, the results show that 1) merely increasing the amount of
instruction data leads to continuous improvement in tasks such as open-ended
generation, 2) in tasks such as math and code, the model performance curve
remains quite flat while increasing data size. We further analyze the possible
causes of these phenomena and propose potential future research directions such
as effectively selecting high-quality training data, scaling base models and
training methods specialized for hard tasks. We will release our training and
evaluation datasets, as well as model checkpoints.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 14:49:37 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Ji",
"Yunjie",
""
],
[
"Deng",
"Yong",
""
],
[
"Gong",
"Yan",
""
],
[
"Peng",
"Yiping",
""
],
[
"Niu",
"Qiang",
""
],
[
"Zhang",
"Lei",
""
],
[
"Ma",
"Baochang",
""
],
[
"Li",
"Xiangang",
""
]
] | TITLE: Exploring the Impact of Instruction Data Scaling on Large Language
Models: An Empirical Study on Real-World Use Cases
ABSTRACT: The success of ChatGPT has recently attracted numerous efforts to replicate
it, with instruction-tuning strategies being a key factor in achieving
remarkable results. Instruction-tuning not only significantly enhances the
model's performance and generalization but also makes the model's generated
results more consistent with human speech patterns. However current research
rarely studies the impact of different amounts of instruction data on model
performance, especially in the real-world use cases. In this paper we explore
the performance of large language models based on instruction tuning across
different scales of instruction data. An evaluation dataset consisting of 12
major online use cases is constructed in the experiment. With Bloomz-7B1-mt as
the base model, the results show that 1) merely increasing the amount of
instruction data leads to continuous improvement in tasks such as open-ended
generation, 2) in tasks such as math and code, the model performance curve
remains quite flat while increasing data size. We further analyze the possible
causes of these phenomena and propose potential future research directions such
as effectively selecting high-quality training data, scaling base models and
training methods specialized for hard tasks. We will release our training and
evaluation datasets, as well as model checkpoints.
| new_dataset | 0.960657 |
2303.14747 | Xiaolong Shen | Xiaolong Shen, Zongxin Yang, Xiaohan Wang, Jianxin Ma, Chang Zhou, Yi
Yang | Global-to-Local Modeling for Video-based 3D Human Pose and Shape
Estimation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video-based 3D human pose and shape estimations are evaluated by intra-frame
accuracy and inter-frame smoothness. Although these two metrics are responsible
for different ranges of temporal consistency, existing state-of-the-art methods
treat them as a unified problem and use monotonous modeling structures (e.g.,
RNN or attention-based block) to design their networks. However, using a single
kind of modeling structure is difficult to balance the learning of short-term
and long-term temporal correlations, and may bias the network to one of them,
leading to undesirable predictions like global location shift, temporal
inconsistency, and insufficient local details. To solve these problems, we
propose to structurally decouple the modeling of long-term and short-term
correlations in an end-to-end framework, Global-to-Local Transformer (GLoT).
First, a global transformer is introduced with a Masked Pose and Shape
Estimation strategy for long-term modeling. The strategy stimulates the global
transformer to learn more inter-frame correlations by randomly masking the
features of several frames. Second, a local transformer is responsible for
exploiting local details on the human mesh and interacting with the global
transformer by leveraging cross-attention. Moreover, a Hierarchical Spatial
Correlation Regressor is further introduced to refine intra-frame estimations
by decoupled global-local representation and implicit kinematic constraints.
Our GLoT surpasses previous state-of-the-art methods with the lowest model
parameters on popular benchmarks, i.e., 3DPW, MPI-INF-3DHP, and Human3.6M.
Codes are available at https://github.com/sxl142/GLoT.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 14:57:49 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Shen",
"Xiaolong",
""
],
[
"Yang",
"Zongxin",
""
],
[
"Wang",
"Xiaohan",
""
],
[
"Ma",
"Jianxin",
""
],
[
"Zhou",
"Chang",
""
],
[
"Yang",
"Yi",
""
]
] | TITLE: Global-to-Local Modeling for Video-based 3D Human Pose and Shape
Estimation
ABSTRACT: Video-based 3D human pose and shape estimations are evaluated by intra-frame
accuracy and inter-frame smoothness. Although these two metrics are responsible
for different ranges of temporal consistency, existing state-of-the-art methods
treat them as a unified problem and use monotonous modeling structures (e.g.,
RNN or attention-based block) to design their networks. However, using a single
kind of modeling structure is difficult to balance the learning of short-term
and long-term temporal correlations, and may bias the network to one of them,
leading to undesirable predictions like global location shift, temporal
inconsistency, and insufficient local details. To solve these problems, we
propose to structurally decouple the modeling of long-term and short-term
correlations in an end-to-end framework, Global-to-Local Transformer (GLoT).
First, a global transformer is introduced with a Masked Pose and Shape
Estimation strategy for long-term modeling. The strategy stimulates the global
transformer to learn more inter-frame correlations by randomly masking the
features of several frames. Second, a local transformer is responsible for
exploiting local details on the human mesh and interacting with the global
transformer by leveraging cross-attention. Moreover, a Hierarchical Spatial
Correlation Regressor is further introduced to refine intra-frame estimations
by decoupled global-local representation and implicit kinematic constraints.
Our GLoT surpasses previous state-of-the-art methods with the lowest model
parameters on popular benchmarks, i.e., 3DPW, MPI-INF-3DHP, and Human3.6M.
Codes are available at https://github.com/sxl142/GLoT.
| no_new_dataset | 0.947186 |
2303.14768 | Bei Gan | Bei Gan, Xiujun Shu, Ruizhi Qiao, Haoqian Wu, Keyu Chen, Hanjun Li, Bo
Ren | Collaborative Noisy Label Cleaner: Learning Scene-aware Trailers for
Multi-modal Highlight Detection in Movies | Accepted to CVPR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Movie highlights stand out of the screenplay for efficient browsing and play
a crucial role on social media platforms. Based on existing efforts, this work
has two observations: (1) For different annotators, labeling highlight has
uncertainty, which leads to inaccurate and time-consuming annotations. (2)
Besides previous supervised or unsupervised settings, some existing video
corpora can be useful, e.g., trailers, but they are often noisy and incomplete
to cover the full highlights. In this work, we study a more practical and
promising setting, i.e., reformulating highlight detection as "learning with
noisy labels". This setting does not require time-consuming manual annotations
and can fully utilize existing abundant video corpora. First, based on movie
trailers, we leverage scene segmentation to obtain complete shots, which are
regarded as noisy labels. Then, we propose a Collaborative noisy Label Cleaner
(CLC) framework to learn from noisy highlight moments. CLC consists of two
modules: augmented cross-propagation (ACP) and multi-modality cleaning (MMC).
The former aims to exploit the closely related audio-visual signals and fuse
them to learn unified multi-modal representations. The latter aims to achieve
cleaner highlight labels by observing the changes in losses among different
modalities. To verify the effectiveness of CLC, we further collect a
large-scale highlight dataset named MovieLights. Comprehensive experiments on
MovieLights and YouTube Highlights datasets demonstrate the effectiveness of
our approach. Code has been made available at:
https://github.com/TencentYoutuResearch/HighlightDetection-CLC
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 16:16:31 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Gan",
"Bei",
""
],
[
"Shu",
"Xiujun",
""
],
[
"Qiao",
"Ruizhi",
""
],
[
"Wu",
"Haoqian",
""
],
[
"Chen",
"Keyu",
""
],
[
"Li",
"Hanjun",
""
],
[
"Ren",
"Bo",
""
]
] | TITLE: Collaborative Noisy Label Cleaner: Learning Scene-aware Trailers for
Multi-modal Highlight Detection in Movies
ABSTRACT: Movie highlights stand out of the screenplay for efficient browsing and play
a crucial role on social media platforms. Based on existing efforts, this work
has two observations: (1) For different annotators, labeling highlight has
uncertainty, which leads to inaccurate and time-consuming annotations. (2)
Besides previous supervised or unsupervised settings, some existing video
corpora can be useful, e.g., trailers, but they are often noisy and incomplete
to cover the full highlights. In this work, we study a more practical and
promising setting, i.e., reformulating highlight detection as "learning with
noisy labels". This setting does not require time-consuming manual annotations
and can fully utilize existing abundant video corpora. First, based on movie
trailers, we leverage scene segmentation to obtain complete shots, which are
regarded as noisy labels. Then, we propose a Collaborative noisy Label Cleaner
(CLC) framework to learn from noisy highlight moments. CLC consists of two
modules: augmented cross-propagation (ACP) and multi-modality cleaning (MMC).
The former aims to exploit the closely related audio-visual signals and fuse
them to learn unified multi-modal representations. The latter aims to achieve
cleaner highlight labels by observing the changes in losses among different
modalities. To verify the effectiveness of CLC, we further collect a
large-scale highlight dataset named MovieLights. Comprehensive experiments on
MovieLights and YouTube Highlights datasets demonstrate the effectiveness of
our approach. Code has been made available at:
https://github.com/TencentYoutuResearch/HighlightDetection-CLC
| new_dataset | 0.958693 |
2303.14770 | Ehsan Shareghi | Thuy-Trang Vu, Xuanli He, Gholamreza Haffari, Ehsan Shareghi | Koala: An Index for Quantifying Overlaps with Pre-training Corpora | Available here: https://koala-index.erc.monash.edu/ | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | In very recent years more attention has been placed on probing the role of
pre-training data in Large Language Models (LLMs) downstream behaviour. Despite
the importance, there is no public tool that supports such analysis of
pre-training corpora at large scale. To help research in this space, we launch
Koala, a searchable index over large pre-training corpora using compressed
suffix arrays with highly efficient compression rate and search support. In its
first release we index the public proportion of OPT 175B pre-training data.
Koala provides a framework to do forensic analysis on the current and future
benchmarks as well as to assess the degree of memorization in the output from
the LLMs. Koala is available for public use at
https://koala-index.erc.monash.edu/.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 16:29:18 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Vu",
"Thuy-Trang",
""
],
[
"He",
"Xuanli",
""
],
[
"Haffari",
"Gholamreza",
""
],
[
"Shareghi",
"Ehsan",
""
]
] | TITLE: Koala: An Index for Quantifying Overlaps with Pre-training Corpora
ABSTRACT: In very recent years more attention has been placed on probing the role of
pre-training data in Large Language Models (LLMs) downstream behaviour. Despite
the importance, there is no public tool that supports such analysis of
pre-training corpora at large scale. To help research in this space, we launch
Koala, a searchable index over large pre-training corpora using compressed
suffix arrays with highly efficient compression rate and search support. In its
first release we index the public proportion of OPT 175B pre-training data.
Koala provides a framework to do forensic analysis on the current and future
benchmarks as well as to assess the degree of memorization in the output from
the LLMs. Koala is available for public use at
https://koala-index.erc.monash.edu/.
| no_new_dataset | 0.893077 |
2303.14777 | Weihua Su | Weihua Sun and Run-An Wang and Zhaonian Zou | Query Generation based on Generative Adversarial Networks | null | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many problems in database systems, such as cardinality estimation, database
testing and optimizer tuning, require a large query load as data. However, it
is often difficult to obtain a large number of real queries from users due to
user privacy restrictions or low frequency of database access. Query generation
is one of the approaches to solve this problem. Existing query generation
methods, such as random generation and template-based generation, do not
consider the relationship between the generated queries and existing queries,
or even generate semantically incorrect queries. In this paper, we propose a
query generation framework based on generative adversarial networks (GAN) to
generate query load that is similar to the given query load. In our framework,
we use a syntax parser to transform the query into a parse tree and traverse
the tree to obtain the sequence of production rules corresponding to the query.
The generator of GAN takes a fixed distribution prior as input and outputs the
query sequence, and the discriminator takes the real query and the fake query
generated by the generator as input and outputs a gradient to guide the
generator learning. In addition, we add context-free grammar and semantic rules
to the generation process, which ensures that the generated queries are
syntactically and semantically correct. We conduct experiments to evaluate our
approach on real-world dataset, which show that our approach can generate new
query loads with a similar distribution to a given query load, and that the
generated queries are syntactically correct with no semantic errors. The
generated query loads are used in downstream task, and the results show a
significant improvement in the models trained with the expanded query loads
using our approach.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 16:55:57 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Sun",
"Weihua",
""
],
[
"Wang",
"Run-An",
""
],
[
"Zou",
"Zhaonian",
""
]
] | TITLE: Query Generation based on Generative Adversarial Networks
ABSTRACT: Many problems in database systems, such as cardinality estimation, database
testing and optimizer tuning, require a large query load as data. However, it
is often difficult to obtain a large number of real queries from users due to
user privacy restrictions or low frequency of database access. Query generation
is one of the approaches to solve this problem. Existing query generation
methods, such as random generation and template-based generation, do not
consider the relationship between the generated queries and existing queries,
or even generate semantically incorrect queries. In this paper, we propose a
query generation framework based on generative adversarial networks (GAN) to
generate query load that is similar to the given query load. In our framework,
we use a syntax parser to transform the query into a parse tree and traverse
the tree to obtain the sequence of production rules corresponding to the query.
The generator of GAN takes a fixed distribution prior as input and outputs the
query sequence, and the discriminator takes the real query and the fake query
generated by the generator as input and outputs a gradient to guide the
generator learning. In addition, we add context-free grammar and semantic rules
to the generation process, which ensures that the generated queries are
syntactically and semantically correct. We conduct experiments to evaluate our
approach on real-world dataset, which show that our approach can generate new
query loads with a similar distribution to a given query load, and that the
generated queries are syntactically correct with no semantic errors. The
generated query loads are used in downstream task, and the results show a
significant improvement in the models trained with the expanded query loads
using our approach.
| no_new_dataset | 0.950778 |
2303.14787 | Taki Hasan Rafi | Taki Hasan Rafi, Faiza Anan Noor, Tahmid Hussain, Dong-Kyu Chae,
Zhaohui Yang | A Generalized Look at Federated Learning: Survey and Perspectives | 9 pages, 2 figures | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning (FL) refers to a distributed machine learning framework
involving learning from several decentralized edge clients without sharing
local dataset. This distributed strategy prevents data leakage and enables
on-device training as it updates the global model based on the local model
updates. Despite offering several advantages, including data privacy and
scalability, FL poses challenges such as statistical and system heterogeneity
of data in federated networks, communication bottlenecks, privacy and security
issues. This survey contains a systematic summarization of previous work,
studies, and experiments on FL and presents a list of possibilities for FL
across a range of applications and use cases. Other than that, various
challenges of implementing FL and promising directions revolving around the
corresponding challenges are provided.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 17:58:11 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Rafi",
"Taki Hasan",
""
],
[
"Noor",
"Faiza Anan",
""
],
[
"Hussain",
"Tahmid",
""
],
[
"Chae",
"Dong-Kyu",
""
],
[
"Yang",
"Zhaohui",
""
]
] | TITLE: A Generalized Look at Federated Learning: Survey and Perspectives
ABSTRACT: Federated learning (FL) refers to a distributed machine learning framework
involving learning from several decentralized edge clients without sharing
local dataset. This distributed strategy prevents data leakage and enables
on-device training as it updates the global model based on the local model
updates. Despite offering several advantages, including data privacy and
scalability, FL poses challenges such as statistical and system heterogeneity
of data in federated networks, communication bottlenecks, privacy and security
issues. This survey contains a systematic summarization of previous work,
studies, and experiments on FL and presents a list of possibilities for FL
across a range of applications and use cases. Other than that, various
challenges of implementing FL and promising directions revolving around the
corresponding challenges are provided.
| no_new_dataset | 0.948537 |
2303.14793 | Ben Vardi | Ben Vardi, Alessandro Torcinovich, Marina Khoroshiltseva, Marcello
Pelillo, Ohad Ben-Shahar | Multi-Phase Relaxation Labeling for Square Jigsaw Puzzle Solving | 10 pages, 7 figures. Published in Proceedings of the 18th
International Joint Conference on Computer Vision, Imaging and Computer
Graphics Theory and Applications - Volume 4 VISAPP: VISAPP, 785-795, 2023 | Published in Proceedings of the 18th International Joint
Conference on Computer Vision, Imaging and Computer Graphics Theory and
Applications - Volume 4 VISAPP: VISAPP, 785-795, 2023 | 10.5220/0011622800003417 | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel method for solving square jigsaw puzzles based on global
optimization. The method is fully automatic, assumes no prior information, and
can handle puzzles with known or unknown piece orientation. At the core of the
optimization process is nonlinear relaxation labeling, a well-founded approach
for deducing global solutions from local constraints, but unlike the classical
scheme here we propose a multi-phase approach that guarantees convergence to
feasible puzzle solutions. Next to the algorithmic novelty, we also present a
new compatibility function for the quantification of the affinity between
adjacent puzzle pieces. Competitive results and the advantage of the
multi-phase approach are demonstrated on standard datasets.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 18:53:51 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Vardi",
"Ben",
""
],
[
"Torcinovich",
"Alessandro",
""
],
[
"Khoroshiltseva",
"Marina",
""
],
[
"Pelillo",
"Marcello",
""
],
[
"Ben-Shahar",
"Ohad",
""
]
] | TITLE: Multi-Phase Relaxation Labeling for Square Jigsaw Puzzle Solving
ABSTRACT: We present a novel method for solving square jigsaw puzzles based on global
optimization. The method is fully automatic, assumes no prior information, and
can handle puzzles with known or unknown piece orientation. At the core of the
optimization process is nonlinear relaxation labeling, a well-founded approach
for deducing global solutions from local constraints, but unlike the classical
scheme here we propose a multi-phase approach that guarantees convergence to
feasible puzzle solutions. Next to the algorithmic novelty, we also present a
new compatibility function for the quantification of the affinity between
adjacent puzzle pieces. Competitive results and the advantage of the
multi-phase approach are demonstrated on standard datasets.
| no_new_dataset | 0.942507 |
2303.14816 | Tian-Zhu Xiang | Zhou Huang, Hang Dai, Tian-Zhu Xiang, Shuo Wang, Huai-Xin Chen, Jie
Qin, Huan Xiong | Feature Shrinkage Pyramid for Camouflaged Object Detection with
Transformers | CVPR 2023. Project webpage at:
https://tzxiang.github.io/project/COD-FSPNet/index.html | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision transformers have recently shown strong global context modeling
capabilities in camouflaged object detection. However, they suffer from two
major limitations: less effective locality modeling and insufficient feature
aggregation in decoders, which are not conducive to camouflaged object
detection that explores subtle cues from indistinguishable backgrounds. To
address these issues, in this paper, we propose a novel transformer-based
Feature Shrinkage Pyramid Network (FSPNet), which aims to hierarchically decode
locality-enhanced neighboring transformer features through progressive
shrinking for camouflaged object detection. Specifically, we propose a nonlocal
token enhancement module (NL-TEM) that employs the non-local mechanism to
interact neighboring tokens and explore graph-based high-order relations within
tokens to enhance local representations of transformers. Moreover, we design a
feature shrinkage decoder (FSD) with adjacent interaction modules (AIM), which
progressively aggregates adjacent transformer features through a layer-bylayer
shrinkage pyramid to accumulate imperceptible but effective cues as much as
possible for object information decoding. Extensive quantitative and
qualitative experiments demonstrate that the proposed model significantly
outperforms the existing 24 competitors on three challenging COD benchmark
datasets under six widely-used evaluation metrics. Our code is publicly
available at https://github.com/ZhouHuang23/FSPNet.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 20:50:58 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Huang",
"Zhou",
""
],
[
"Dai",
"Hang",
""
],
[
"Xiang",
"Tian-Zhu",
""
],
[
"Wang",
"Shuo",
""
],
[
"Chen",
"Huai-Xin",
""
],
[
"Qin",
"Jie",
""
],
[
"Xiong",
"Huan",
""
]
] | TITLE: Feature Shrinkage Pyramid for Camouflaged Object Detection with
Transformers
ABSTRACT: Vision transformers have recently shown strong global context modeling
capabilities in camouflaged object detection. However, they suffer from two
major limitations: less effective locality modeling and insufficient feature
aggregation in decoders, which are not conducive to camouflaged object
detection that explores subtle cues from indistinguishable backgrounds. To
address these issues, in this paper, we propose a novel transformer-based
Feature Shrinkage Pyramid Network (FSPNet), which aims to hierarchically decode
locality-enhanced neighboring transformer features through progressive
shrinking for camouflaged object detection. Specifically, we propose a nonlocal
token enhancement module (NL-TEM) that employs the non-local mechanism to
interact neighboring tokens and explore graph-based high-order relations within
tokens to enhance local representations of transformers. Moreover, we design a
feature shrinkage decoder (FSD) with adjacent interaction modules (AIM), which
progressively aggregates adjacent transformer features through a layer-bylayer
shrinkage pyramid to accumulate imperceptible but effective cues as much as
possible for object information decoding. Extensive quantitative and
qualitative experiments demonstrate that the proposed model significantly
outperforms the existing 24 competitors on three challenging COD benchmark
datasets under six widely-used evaluation metrics. Our code is publicly
available at https://github.com/ZhouHuang23/FSPNet.
| no_new_dataset | 0.945901 |
2303.14817 | Yitian Zhang | Yitian Zhang, Yue Bai, Chang Liu, Huan Wang, Sheng Li, Yun Fu | Frame Flexible Network | Accepted by CVPR2023 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing video recognition algorithms always conduct different training
pipelines for inputs with different frame numbers, which requires repetitive
training operations and multiplying storage costs. If we evaluate the model
using other frames which are not used in training, we observe the performance
will drop significantly (see Fig.1), which is summarized as Temporal Frequency
Deviation phenomenon. To fix this issue, we propose a general framework, named
Frame Flexible Network (FFN), which not only enables the model to be evaluated
at different frames to adjust its computation, but also reduces the memory
costs of storing multiple models significantly. Concretely, FFN integrates
several sets of training sequences, involves Multi-Frequency Alignment (MFAL)
to learn temporal frequency invariant representations, and leverages
Multi-Frequency Adaptation (MFAD) to further strengthen the representation
abilities. Comprehensive empirical validations using various architectures and
popular benchmarks solidly demonstrate the effectiveness and generalization of
FFN (e.g., 7.08/5.15/2.17% performance gain at Frame 4/8/16 on
Something-Something V1 dataset over Uniformer). Code is available at
https://github.com/BeSpontaneous/FFN.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 20:51:35 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Yitian",
""
],
[
"Bai",
"Yue",
""
],
[
"Liu",
"Chang",
""
],
[
"Wang",
"Huan",
""
],
[
"Li",
"Sheng",
""
],
[
"Fu",
"Yun",
""
]
] | TITLE: Frame Flexible Network
ABSTRACT: Existing video recognition algorithms always conduct different training
pipelines for inputs with different frame numbers, which requires repetitive
training operations and multiplying storage costs. If we evaluate the model
using other frames which are not used in training, we observe the performance
will drop significantly (see Fig.1), which is summarized as Temporal Frequency
Deviation phenomenon. To fix this issue, we propose a general framework, named
Frame Flexible Network (FFN), which not only enables the model to be evaluated
at different frames to adjust its computation, but also reduces the memory
costs of storing multiple models significantly. Concretely, FFN integrates
several sets of training sequences, involves Multi-Frequency Alignment (MFAL)
to learn temporal frequency invariant representations, and leverages
Multi-Frequency Adaptation (MFAD) to further strengthen the representation
abilities. Comprehensive empirical validations using various architectures and
popular benchmarks solidly demonstrate the effectiveness and generalization of
FFN (e.g., 7.08/5.15/2.17% performance gain at Frame 4/8/16 on
Something-Something V1 dataset over Uniformer). Code is available at
https://github.com/BeSpontaneous/FFN.
| no_new_dataset | 0.949248 |
2303.14828 | Dina Bashkirova | Dina Bashkirova, Samarth Mishra, Diala Lteif, Piotr Teterwak, Donghyun
Kim, Fadi Alladkani, James Akl, Berk Calli, Sarah Adel Bargal, Kate Saenko,
Daehan Kim, Minseok Seo, YoungJin Jeon, Dong-Geol Choi, Shahaf Ettedgui, Raja
Giryes, Shady Abu-Hussein, Binhui Xie, Shuang Li | VisDA 2022 Challenge: Domain Adaptation for Industrial Waste Sorting | Proceedings of Machine Learning Research | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Label-efficient and reliable semantic segmentation is essential for many
real-life applications, especially for industrial settings with high visual
diversity, such as waste sorting. In industrial waste sorting, one of the
biggest challenges is the extreme diversity of the input stream depending on
factors like the location of the sorting facility, the equipment available in
the facility, and the time of year, all of which significantly impact the
composition and visual appearance of the waste stream. These changes in the
data are called ``visual domains'', and label-efficient adaptation of models to
such domains is needed for successful semantic segmentation of industrial
waste. To test the abilities of computer vision models on this task, we present
the VisDA 2022 Challenge on Domain Adaptation for Industrial Waste Sorting. Our
challenge incorporates a fully-annotated waste sorting dataset, ZeroWaste,
collected from two real material recovery facilities in different locations and
seasons, as well as a novel procedurally generated synthetic waste sorting
dataset, SynthWaste. In this competition, we aim to answer two questions: 1)
can we leverage domain adaptation techniques to minimize the domain gap? and 2)
can synthetic data augmentation improve performance on this task and help adapt
to changing data distributions? The results of the competition show that
industrial waste detection poses a real domain adaptation problem, that domain
generalization techniques such as augmentations, ensembling, etc., improve the
overall performance on the unlabeled target domain examples, and that
leveraging synthetic data effectively remains an open problem. See
https://ai.bu.edu/visda-2022/
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 21:38:38 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Bashkirova",
"Dina",
""
],
[
"Mishra",
"Samarth",
""
],
[
"Lteif",
"Diala",
""
],
[
"Teterwak",
"Piotr",
""
],
[
"Kim",
"Donghyun",
""
],
[
"Alladkani",
"Fadi",
""
],
[
"Akl",
"James",
""
],
[
"Calli",
"Berk",
""
],
[
"Bargal",
"Sarah Adel",
""
],
[
"Saenko",
"Kate",
""
],
[
"Kim",
"Daehan",
""
],
[
"Seo",
"Minseok",
""
],
[
"Jeon",
"YoungJin",
""
],
[
"Choi",
"Dong-Geol",
""
],
[
"Ettedgui",
"Shahaf",
""
],
[
"Giryes",
"Raja",
""
],
[
"Abu-Hussein",
"Shady",
""
],
[
"Xie",
"Binhui",
""
],
[
"Li",
"Shuang",
""
]
] | TITLE: VisDA 2022 Challenge: Domain Adaptation for Industrial Waste Sorting
ABSTRACT: Label-efficient and reliable semantic segmentation is essential for many
real-life applications, especially for industrial settings with high visual
diversity, such as waste sorting. In industrial waste sorting, one of the
biggest challenges is the extreme diversity of the input stream depending on
factors like the location of the sorting facility, the equipment available in
the facility, and the time of year, all of which significantly impact the
composition and visual appearance of the waste stream. These changes in the
data are called ``visual domains'', and label-efficient adaptation of models to
such domains is needed for successful semantic segmentation of industrial
waste. To test the abilities of computer vision models on this task, we present
the VisDA 2022 Challenge on Domain Adaptation for Industrial Waste Sorting. Our
challenge incorporates a fully-annotated waste sorting dataset, ZeroWaste,
collected from two real material recovery facilities in different locations and
seasons, as well as a novel procedurally generated synthetic waste sorting
dataset, SynthWaste. In this competition, we aim to answer two questions: 1)
can we leverage domain adaptation techniques to minimize the domain gap? and 2)
can synthetic data augmentation improve performance on this task and help adapt
to changing data distributions? The results of the competition show that
industrial waste detection poses a real domain adaptation problem, that domain
generalization techniques such as augmentations, ensembling, etc., improve the
overall performance on the unlabeled target domain examples, and that
leveraging synthetic data effectively remains an open problem. See
https://ai.bu.edu/visda-2022/
| new_dataset | 0.968974 |
2303.14840 | Patrick Ruhkamp | HyunJun Jung, Patrick Ruhkamp, Guangyao Zhai, Nikolas Brasch, Yitong
Li, Yannick Verdie, Jifei Song, Yiren Zhou, Anil Armagan, Slobodan Ilic, Ales
Leonardis, Nassir Navab, Benjamin Busam | On the Importance of Accurate Geometry Data for Dense 3D Vision Tasks | Accepted at CVPR 2023, Main Paper + Supp. Mat. arXiv admin note:
substantial text overlap with arXiv:2205.04565 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Learning-based methods to solve dense 3D vision problems typically train on
3D sensor data. The respectively used principle of measuring distances provides
advantages and drawbacks. These are typically not compared nor discussed in the
literature due to a lack of multi-modal datasets. Texture-less regions are
problematic for structure from motion and stereo, reflective material poses
issues for active sensing, and distances for translucent objects are intricate
to measure with existing hardware. Training on inaccurate or corrupt data
induces model bias and hampers generalisation capabilities. These effects
remain unnoticed if the sensor measurement is considered as ground truth during
the evaluation. This paper investigates the effect of sensor errors for the
dense 3D vision tasks of depth estimation and reconstruction. We rigorously
show the significant impact of sensor characteristics on the learned
predictions and notice generalisation issues arising from various technologies
in everyday household environments. For evaluation, we introduce a carefully
designed dataset\footnote{dataset available at
https://github.com/Junggy/HAMMER-dataset} comprising measurements from
commodity sensors, namely D-ToF, I-ToF, passive/active stereo, and monocular
RGB+P. Our study quantifies the considerable sensor noise impact and paves the
way to improved dense vision estimates and targeted data fusion.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 22:32:44 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Jung",
"HyunJun",
""
],
[
"Ruhkamp",
"Patrick",
""
],
[
"Zhai",
"Guangyao",
""
],
[
"Brasch",
"Nikolas",
""
],
[
"Li",
"Yitong",
""
],
[
"Verdie",
"Yannick",
""
],
[
"Song",
"Jifei",
""
],
[
"Zhou",
"Yiren",
""
],
[
"Armagan",
"Anil",
""
],
[
"Ilic",
"Slobodan",
""
],
[
"Leonardis",
"Ales",
""
],
[
"Navab",
"Nassir",
""
],
[
"Busam",
"Benjamin",
""
]
] | TITLE: On the Importance of Accurate Geometry Data for Dense 3D Vision Tasks
ABSTRACT: Learning-based methods to solve dense 3D vision problems typically train on
3D sensor data. The respectively used principle of measuring distances provides
advantages and drawbacks. These are typically not compared nor discussed in the
literature due to a lack of multi-modal datasets. Texture-less regions are
problematic for structure from motion and stereo, reflective material poses
issues for active sensing, and distances for translucent objects are intricate
to measure with existing hardware. Training on inaccurate or corrupt data
induces model bias and hampers generalisation capabilities. These effects
remain unnoticed if the sensor measurement is considered as ground truth during
the evaluation. This paper investigates the effect of sensor errors for the
dense 3D vision tasks of depth estimation and reconstruction. We rigorously
show the significant impact of sensor characteristics on the learned
predictions and notice generalisation issues arising from various technologies
in everyday household environments. For evaluation, we introduce a carefully
designed dataset\footnote{dataset available at
https://github.com/Junggy/HAMMER-dataset} comprising measurements from
commodity sensors, namely D-ToF, I-ToF, passive/active stereo, and monocular
RGB+P. Our study quantifies the considerable sensor noise impact and paves the
way to improved dense vision estimates and targeted data fusion.
| new_dataset | 0.962108 |
2303.14859 | Junchi Yu | Junchi Yu and Jian Liang and Ran He | Mind the Label Shift of Augmentation-based Graph OOD Generalization | Accepted to CVPR 2023. arXiv admin note: text overlap with
arXiv:2206.09345 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Out-of-distribution (OOD) generalization is an important issue for Graph
Neural Networks (GNNs). Recent works employ different graph editions to
generate augmented environments and learn an invariant GNN for generalization.
However, the label shift usually occurs in augmentation since graph structural
edition inevitably alters the graph label. This brings inconsistent predictive
relationships among augmented environments, which is harmful to generalization.
To address this issue, we propose \textbf{LiSA}, which generates
label-invariant augmentations to facilitate graph OOD generalization. Instead
of resorting to graph editions, LiSA exploits \textbf{L}abel-\textbf{i}nvariant
\textbf{S}ubgraphs of the training graphs to construct \textbf{A}ugmented
environments. Specifically, LiSA first designs the variational subgraph
generators to extract locally predictive patterns and construct multiple
label-invariant subgraphs efficiently. Then, the subgraphs produced by
different generators are collected to build different augmented environments.
To promote diversity among augmented environments, LiSA further introduces a
tractable energy-based regularization to enlarge pair-wise distances between
the distributions of environments. In this manner, LiSA generates diverse
augmented environments with a consistent predictive relationship and
facilitates learning an invariant GNN. Extensive experiments on node-level and
graph-level OOD benchmarks show that LiSA achieves impressive generalization
performance with different GNN backbones. Code is available on
\url{https://github.com/Samyu0304/LiSA}.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 00:08:45 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yu",
"Junchi",
""
],
[
"Liang",
"Jian",
""
],
[
"He",
"Ran",
""
]
] | TITLE: Mind the Label Shift of Augmentation-based Graph OOD Generalization
ABSTRACT: Out-of-distribution (OOD) generalization is an important issue for Graph
Neural Networks (GNNs). Recent works employ different graph editions to
generate augmented environments and learn an invariant GNN for generalization.
However, the label shift usually occurs in augmentation since graph structural
edition inevitably alters the graph label. This brings inconsistent predictive
relationships among augmented environments, which is harmful to generalization.
To address this issue, we propose \textbf{LiSA}, which generates
label-invariant augmentations to facilitate graph OOD generalization. Instead
of resorting to graph editions, LiSA exploits \textbf{L}abel-\textbf{i}nvariant
\textbf{S}ubgraphs of the training graphs to construct \textbf{A}ugmented
environments. Specifically, LiSA first designs the variational subgraph
generators to extract locally predictive patterns and construct multiple
label-invariant subgraphs efficiently. Then, the subgraphs produced by
different generators are collected to build different augmented environments.
To promote diversity among augmented environments, LiSA further introduces a
tractable energy-based regularization to enlarge pair-wise distances between
the distributions of environments. In this manner, LiSA generates diverse
augmented environments with a consistent predictive relationship and
facilitates learning an invariant GNN. Extensive experiments on node-level and
graph-level OOD benchmarks show that LiSA achieves impressive generalization
performance with different GNN backbones. Code is available on
\url{https://github.com/Samyu0304/LiSA}.
| no_new_dataset | 0.948965 |
2303.14885 | Karren Yang | Karren Yang, Ting-Yao Hu, Jen-Hao Rick Chang, Hema Swetha Koppula,
Oncel Tuzel | Text is All You Need: Personalizing ASR Models using Controllable Speech
Synthesis | ICASSP 2023 | null | null | null | eess.AS cs.LG cs.SD | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Adapting generic speech recognition models to specific individuals is a
challenging problem due to the scarcity of personalized data. Recent works have
proposed boosting the amount of training data using personalized text-to-speech
synthesis. Here, we ask two fundamental questions about this strategy: when is
synthetic data effective for personalization, and why is it effective in those
cases? To address the first question, we adapt a state-of-the-art automatic
speech recognition (ASR) model to target speakers from four benchmark datasets
representative of different speaker types. We show that ASR personalization
with synthetic data is effective in all cases, but particularly when (i) the
target speaker is underrepresented in the global data, and (ii) the capacity of
the global model is limited. To address the second question of why personalized
synthetic data is effective, we use controllable speech synthesis to generate
speech with varied styles and content. Surprisingly, we find that the text
content of the synthetic data, rather than style, is important for speaker
adaptation. These results lead us to propose a data selection strategy for ASR
personalization based on speech content.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 02:50:02 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yang",
"Karren",
""
],
[
"Hu",
"Ting-Yao",
""
],
[
"Chang",
"Jen-Hao Rick",
""
],
[
"Koppula",
"Hema Swetha",
""
],
[
"Tuzel",
"Oncel",
""
]
] | TITLE: Text is All You Need: Personalizing ASR Models using Controllable Speech
Synthesis
ABSTRACT: Adapting generic speech recognition models to specific individuals is a
challenging problem due to the scarcity of personalized data. Recent works have
proposed boosting the amount of training data using personalized text-to-speech
synthesis. Here, we ask two fundamental questions about this strategy: when is
synthetic data effective for personalization, and why is it effective in those
cases? To address the first question, we adapt a state-of-the-art automatic
speech recognition (ASR) model to target speakers from four benchmark datasets
representative of different speaker types. We show that ASR personalization
with synthetic data is effective in all cases, but particularly when (i) the
target speaker is underrepresented in the global data, and (ii) the capacity of
the global model is limited. To address the second question of why personalized
synthetic data is effective, we use controllable speech synthesis to generate
speech with varied styles and content. Surprisingly, we find that the text
content of the synthetic data, rather than style, is important for speaker
adaptation. These results lead us to propose a data selection strategy for ASR
personalization based on speech content.
| no_new_dataset | 0.953188 |
2303.14888 | Ruoqi Yin | Ruoqi Yin, Jianqin Yin | Global Relation Modeling and Refinement for Bottom-Up Human Pose
Estimation | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we concern on the bottom-up paradigm in multi-person pose
estimation (MPPE). Most previous bottom-up methods try to consider the relation
of instances to identify different body parts during the post processing, while
ignoring to model the relation among instances or environment in the feature
learning process. In addition, most existing works adopt the operations of
upsampling and downsampling. During the sampling process, there will be a
problem of misalignment with the source features, resulting in deviations in
the keypoint features learned by the model.
To overcome the above limitations, we propose a convolutional neural network
for bottom-up human pose estimation. It invovles two basic modules: (i) Global
Relation Modeling (GRM) module globally learns relation (e.g., environment
context, instance interactive information) among region of image by fusing
multiple stages features in the feature learning process. It combines with the
spatial-channel attention mechanism, which focuses on achieving adaptability in
spatial and channel dimensions. (ii) Multi-branch Feature Align (MFA) module
aggregates features from multiple branches to align fused feature and obtain
refined local keypoint representation. Our model has the ability to focus on
different granularity from local to global regions, which significantly boosts
the performance of the multi-person pose estimation. Our results on the COCO
and CrowdPose datasets demonstrate that it is an efficient framework for
multi-person pose estimation.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 02:54:08 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Yin",
"Ruoqi",
""
],
[
"Yin",
"Jianqin",
""
]
] | TITLE: Global Relation Modeling and Refinement for Bottom-Up Human Pose
Estimation
ABSTRACT: In this paper, we concern on the bottom-up paradigm in multi-person pose
estimation (MPPE). Most previous bottom-up methods try to consider the relation
of instances to identify different body parts during the post processing, while
ignoring to model the relation among instances or environment in the feature
learning process. In addition, most existing works adopt the operations of
upsampling and downsampling. During the sampling process, there will be a
problem of misalignment with the source features, resulting in deviations in
the keypoint features learned by the model.
To overcome the above limitations, we propose a convolutional neural network
for bottom-up human pose estimation. It invovles two basic modules: (i) Global
Relation Modeling (GRM) module globally learns relation (e.g., environment
context, instance interactive information) among region of image by fusing
multiple stages features in the feature learning process. It combines with the
spatial-channel attention mechanism, which focuses on achieving adaptability in
spatial and channel dimensions. (ii) Multi-branch Feature Align (MFA) module
aggregates features from multiple branches to align fused feature and obtain
refined local keypoint representation. Our model has the ability to focus on
different granularity from local to global regions, which significantly boosts
the performance of the multi-person pose estimation. Our results on the COCO
and CrowdPose datasets demonstrate that it is an efficient framework for
multi-person pose estimation.
| no_new_dataset | 0.946001 |
2303.14894 | Md Solimul Chowdhury | Md Solimul Chowdhury and Cayden R. Codel and Marijn J.H. Heule | A Linear Weight Transfer Rule for Local Search | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | The Divide and Distribute Fixed Weights algorithm (ddfw) is a dynamic local
search SAT-solving algorithm that transfers weight from satisfied to falsified
clauses in local minima. ddfw is remarkably effective on several hard
combinatorial instances. Yet, despite its success, it has received little study
since its debut in 2005. In this paper, we propose three modifications to the
base algorithm: a linear weight transfer method that moves a dynamic amount of
weight between clauses in local minima, an adjustment to how satisfied clauses
are chosen in local minima to give weight, and a weighted-random method of
selecting variables to flip. We implemented our modifications to ddfw on top of
the solver yalsat. Our experiments show that our modifications boost the
performance compared to the original ddfw algorithm on multiple benchmarks,
including those from the past three years of SAT competitions. Moreover, our
improved solver exclusively solves hard combinatorial instances that refute a
conjecture on the lower bound of two Van der Waerden numbers set forth by Ahmed
et al. (2014), and it performs well on a hard graph-coloring instance that has
been open for over three decades.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 03:06:34 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chowdhury",
"Md Solimul",
""
],
[
"Codel",
"Cayden R.",
""
],
[
"Heule",
"Marijn J. H.",
""
]
] | TITLE: A Linear Weight Transfer Rule for Local Search
ABSTRACT: The Divide and Distribute Fixed Weights algorithm (ddfw) is a dynamic local
search SAT-solving algorithm that transfers weight from satisfied to falsified
clauses in local minima. ddfw is remarkably effective on several hard
combinatorial instances. Yet, despite its success, it has received little study
since its debut in 2005. In this paper, we propose three modifications to the
base algorithm: a linear weight transfer method that moves a dynamic amount of
weight between clauses in local minima, an adjustment to how satisfied clauses
are chosen in local minima to give weight, and a weighted-random method of
selecting variables to flip. We implemented our modifications to ddfw on top of
the solver yalsat. Our experiments show that our modifications boost the
performance compared to the original ddfw algorithm on multiple benchmarks,
including those from the past three years of SAT competitions. Moreover, our
improved solver exclusively solves hard combinatorial instances that refute a
conjecture on the lower bound of two Van der Waerden numbers set forth by Ahmed
et al. (2014), and it performs well on a hard graph-coloring instance that has
been open for over three decades.
| no_new_dataset | 0.945901 |
2303.14895 | Xiaofan Li | Xiaofan Li, Xuan Li, Guangfa Lv, Yongzheng Zhang and Fengyu Wang | Directed Greybox Fuzzing with Stepwise Constraint Focusing | null | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic data flow analysis has been widely used to guide greybox fuzzing.
However, traditional dynamic data flow analysis tends to go astray in the
massive path tracking and requires to process a large volume of data, resulting
in low efficiency in reaching the target location. In this paper, we propose a
directed greybox fuzzer based on dynamic constraint filtering and focusing
(CONFF). First, all path constraints are tracked, and those with high priority
are filtered as the next solution targets. Next, focusing on a single path
constraint to be satisfied, we obtain its data condition and probe the mapping
relationship between it and the input bytes through multi-byte mapping and
single-byte mapping. Finally, various mutation strategies are utilized to solve
the path constraint currently focused on, and the target location of the
program is gradually approached through path selection. The CONFF fuzzer can
reach a specific location faster in the target program, thus efficiently
triggering the crash. We designed and implemented a prototype of the CONFF
fuzzer and evaluated it with the LAVA-1 dataset and some real-world
vulnerabilities. The results show that the CONFF fuzzer can reproduce crashes
on the LAVA-1 dataset and most of the real-world vulnerabilities. For most
vulnerabilities, the CONFF fuzzer reproduced the crashes with significantly
reduced time compared to state-of-the-art fuzzers. On average, the CONFF fuzzer
was 23.7x faster than the state-of-the-art code coverage-based fuzzer Angora
and 27.3x faster than the classical directed greybox fuzzer AFLGo.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 03:09:32 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Li",
"Xiaofan",
""
],
[
"Li",
"Xuan",
""
],
[
"Lv",
"Guangfa",
""
],
[
"Zhang",
"Yongzheng",
""
],
[
"Wang",
"Fengyu",
""
]
] | TITLE: Directed Greybox Fuzzing with Stepwise Constraint Focusing
ABSTRACT: Dynamic data flow analysis has been widely used to guide greybox fuzzing.
However, traditional dynamic data flow analysis tends to go astray in the
massive path tracking and requires to process a large volume of data, resulting
in low efficiency in reaching the target location. In this paper, we propose a
directed greybox fuzzer based on dynamic constraint filtering and focusing
(CONFF). First, all path constraints are tracked, and those with high priority
are filtered as the next solution targets. Next, focusing on a single path
constraint to be satisfied, we obtain its data condition and probe the mapping
relationship between it and the input bytes through multi-byte mapping and
single-byte mapping. Finally, various mutation strategies are utilized to solve
the path constraint currently focused on, and the target location of the
program is gradually approached through path selection. The CONFF fuzzer can
reach a specific location faster in the target program, thus efficiently
triggering the crash. We designed and implemented a prototype of the CONFF
fuzzer and evaluated it with the LAVA-1 dataset and some real-world
vulnerabilities. The results show that the CONFF fuzzer can reproduce crashes
on the LAVA-1 dataset and most of the real-world vulnerabilities. For most
vulnerabilities, the CONFF fuzzer reproduced the crashes with significantly
reduced time compared to state-of-the-art fuzzers. On average, the CONFF fuzzer
was 23.7x faster than the state-of-the-art code coverage-based fuzzer Angora
and 27.3x faster than the classical directed greybox fuzzer AFLGo.
| no_new_dataset | 0.949153 |
2303.14898 | Wang Ruijie | Ruijie Wang, Zheng Li, Jingfeng Yang, Tianyu Cao, Chao Zhang, Bing Yin
and Tarek Abdelzaher | Mutually-paced Knowledge Distillation for Cross-lingual Temporal
Knowledge Graph Reasoning | This paper is accepted by The Web Conference 2023 | null | null | null | cs.LG cs.AI cs.CL cs.SI | http://creativecommons.org/licenses/by/4.0/ | This paper investigates cross-lingual temporal knowledge graph reasoning
problem, which aims to facilitate reasoning on Temporal Knowledge Graphs (TKGs)
in low-resource languages by transfering knowledge from TKGs in high-resource
ones. The cross-lingual distillation ability across TKGs becomes increasingly
crucial, in light of the unsatisfying performance of existing reasoning methods
on those severely incomplete TKGs, especially in low-resource languages.
However, it poses tremendous challenges in two aspects. First, the
cross-lingual alignments, which serve as bridges for knowledge transfer, are
usually too scarce to transfer sufficient knowledge between two TKGs. Second,
temporal knowledge discrepancy of the aligned entities, especially when
alignments are unreliable, can mislead the knowledge distillation process. We
correspondingly propose a mutually-paced knowledge distillation model MP-KD,
where a teacher network trained on a source TKG can guide the training of a
student network on target TKGs with an alignment module. Concretely, to deal
with the scarcity issue, MP-KD generates pseudo alignments between TKGs based
on the temporal information extracted by our representation module. To maximize
the efficacy of knowledge transfer and control the noise caused by the temporal
knowledge discrepancy, we enhance MP-KD with a temporal cross-lingual attention
mechanism to dynamically estimate the alignment strength. The two procedures
are mutually paced along with model training. Extensive experiments on twelve
cross-lingual TKG transfer tasks in the EventKG benchmark demonstrate the
effectiveness of the proposed MP-KD method.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 03:15:27 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Ruijie",
""
],
[
"Li",
"Zheng",
""
],
[
"Yang",
"Jingfeng",
""
],
[
"Cao",
"Tianyu",
""
],
[
"Zhang",
"Chao",
""
],
[
"Yin",
"Bing",
""
],
[
"Abdelzaher",
"Tarek",
""
]
] | TITLE: Mutually-paced Knowledge Distillation for Cross-lingual Temporal
Knowledge Graph Reasoning
ABSTRACT: This paper investigates cross-lingual temporal knowledge graph reasoning
problem, which aims to facilitate reasoning on Temporal Knowledge Graphs (TKGs)
in low-resource languages by transfering knowledge from TKGs in high-resource
ones. The cross-lingual distillation ability across TKGs becomes increasingly
crucial, in light of the unsatisfying performance of existing reasoning methods
on those severely incomplete TKGs, especially in low-resource languages.
However, it poses tremendous challenges in two aspects. First, the
cross-lingual alignments, which serve as bridges for knowledge transfer, are
usually too scarce to transfer sufficient knowledge between two TKGs. Second,
temporal knowledge discrepancy of the aligned entities, especially when
alignments are unreliable, can mislead the knowledge distillation process. We
correspondingly propose a mutually-paced knowledge distillation model MP-KD,
where a teacher network trained on a source TKG can guide the training of a
student network on target TKGs with an alignment module. Concretely, to deal
with the scarcity issue, MP-KD generates pseudo alignments between TKGs based
on the temporal information extracted by our representation module. To maximize
the efficacy of knowledge transfer and control the noise caused by the temporal
knowledge discrepancy, we enhance MP-KD with a temporal cross-lingual attention
mechanism to dynamically estimate the alignment strength. The two procedures
are mutually paced along with model training. Extensive experiments on twelve
cross-lingual TKG transfer tasks in the EventKG benchmark demonstrate the
effectiveness of the proposed MP-KD method.
| no_new_dataset | 0.945951 |
2303.14901 | Ryo Toda | Ryo Toda, Hayato Itoh, Masahiro Oda, Yuichiro Hayashi, Yoshito Otake,
Masahiro Hashimoto, Toshiaki Akashi, Shigeki Aoki, Kensaku Mori | Identifying Suspicious Regions of Covid-19 by Abnormality-Sensitive
Activation Mapping | 10 pages, 3 figures | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a fully-automated method for the identification of
suspicious regions of a coronavirus disease (COVID-19) on chest CT volumes. One
major role of chest CT scanning in COVID-19 diagnoses is identification of an
inflammation particular to the disease. This task is generally performed by
radiologists through an interpretation of the CT volumes, however, because of
the heavy workload, an automatic analysis method using a computer is desired.
Most computer-aided diagnosis studies have addressed only a portion of the
elements necessary for the identification. In this work, we realize the
identification method through a classification task by using a 2.5-dimensional
CNN with three-dimensional attention mechanisms. We visualize the suspicious
regions by applying a backpropagation based on positive gradients to
attention-weighted features. We perform experiments on an in-house dataset and
two public datasets to reveal the generalization ability of the proposed
method. The proposed architecture achieved AUCs of over 0.900 for all the
datasets, and mean sensitivity $0.853 \pm 0.036$ and specificity $0.870 \pm
0.040$. The method can also identify notable lesions pointed out in the
radiology report as suspicious regions.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 03:22:25 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Toda",
"Ryo",
""
],
[
"Itoh",
"Hayato",
""
],
[
"Oda",
"Masahiro",
""
],
[
"Hayashi",
"Yuichiro",
""
],
[
"Otake",
"Yoshito",
""
],
[
"Hashimoto",
"Masahiro",
""
],
[
"Akashi",
"Toshiaki",
""
],
[
"Aoki",
"Shigeki",
""
],
[
"Mori",
"Kensaku",
""
]
] | TITLE: Identifying Suspicious Regions of Covid-19 by Abnormality-Sensitive
Activation Mapping
ABSTRACT: This paper presents a fully-automated method for the identification of
suspicious regions of a coronavirus disease (COVID-19) on chest CT volumes. One
major role of chest CT scanning in COVID-19 diagnoses is identification of an
inflammation particular to the disease. This task is generally performed by
radiologists through an interpretation of the CT volumes, however, because of
the heavy workload, an automatic analysis method using a computer is desired.
Most computer-aided diagnosis studies have addressed only a portion of the
elements necessary for the identification. In this work, we realize the
identification method through a classification task by using a 2.5-dimensional
CNN with three-dimensional attention mechanisms. We visualize the suspicious
regions by applying a backpropagation based on positive gradients to
attention-weighted features. We perform experiments on an in-house dataset and
two public datasets to reveal the generalization ability of the proposed
method. The proposed architecture achieved AUCs of over 0.900 for all the
datasets, and mean sensitivity $0.853 \pm 0.036$ and specificity $0.870 \pm
0.040$. The method can also identify notable lesions pointed out in the
radiology report as suspicious regions.
| no_new_dataset | 0.941277 |
2303.14920 | Christopher McMaster | Christopher McMaster, David FL Liew, Douglas EV Pires | Adapting Pretrained Language Models for Solving Tabular Prediction
Problems in the Electronic Health Record | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | We propose an approach for adapting the DeBERTa model for electronic health
record (EHR) tasks using domain adaptation. We pretrain a small DeBERTa model
on a dataset consisting of MIMIC-III discharge summaries, clinical notes,
radiology reports, and PubMed abstracts. We compare this model's performance
with a DeBERTa model pre-trained on clinical texts from our institutional EHR
(MeDeBERTa) and an XGBoost model. We evaluate performance on three benchmark
tasks for emergency department outcomes using the MIMIC-IV-ED dataset. We
preprocess the data to convert it into text format and generate four versions
of the original datasets to compare data processing and data inclusion. The
results show that our proposed approach outperforms the alternative models on
two of three tasks (p<0.001) and matches performance on the third task, with
the use of descriptive columns improving performance over the original column
names.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 05:34:19 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"McMaster",
"Christopher",
""
],
[
"Liew",
"David FL",
""
],
[
"Pires",
"Douglas EV",
""
]
] | TITLE: Adapting Pretrained Language Models for Solving Tabular Prediction
Problems in the Electronic Health Record
ABSTRACT: We propose an approach for adapting the DeBERTa model for electronic health
record (EHR) tasks using domain adaptation. We pretrain a small DeBERTa model
on a dataset consisting of MIMIC-III discharge summaries, clinical notes,
radiology reports, and PubMed abstracts. We compare this model's performance
with a DeBERTa model pre-trained on clinical texts from our institutional EHR
(MeDeBERTa) and an XGBoost model. We evaluate performance on three benchmark
tasks for emergency department outcomes using the MIMIC-IV-ED dataset. We
preprocess the data to convert it into text format and generate four versions
of the original datasets to compare data processing and data inclusion. The
results show that our proposed approach outperforms the alternative models on
two of three tasks (p<0.001) and matches performance on the third task, with
the use of descriptive columns improving performance over the original column
names.
| no_new_dataset | 0.936285 |
2303.14922 | Xingbin Liu | Xingbin Liu, Huafeng Kuang, Xianming Lin, Yongjian Wu, Rongrong Ji | CAT:Collaborative Adversarial Training | Tech report | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adversarial training can improve the robustness of neural networks. Previous
methods focus on a single adversarial training strategy and do not consider the
model property trained by different strategies. By revisiting the previous
methods, we find different adversarial training methods have distinct
robustness for sample instances. For example, a sample instance can be
correctly classified by a model trained using standard adversarial training
(AT) but not by a model trained using TRADES, and vice versa. Based on this
observation, we propose a collaborative adversarial training framework to
improve the robustness of neural networks. Specifically, we use different
adversarial training methods to train robust models and let models interact
with their knowledge during the training process. Collaborative Adversarial
Training (CAT) can improve both robustness and accuracy. Extensive experiments
on various networks and datasets validate the effectiveness of our method. CAT
achieves state-of-the-art adversarial robustness without using any additional
data on CIFAR-10 under the Auto-Attack benchmark. Code is available at
https://github.com/liuxingbin/CAT.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 05:37:43 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Liu",
"Xingbin",
""
],
[
"Kuang",
"Huafeng",
""
],
[
"Lin",
"Xianming",
""
],
[
"Wu",
"Yongjian",
""
],
[
"Ji",
"Rongrong",
""
]
] | TITLE: CAT:Collaborative Adversarial Training
ABSTRACT: Adversarial training can improve the robustness of neural networks. Previous
methods focus on a single adversarial training strategy and do not consider the
model property trained by different strategies. By revisiting the previous
methods, we find different adversarial training methods have distinct
robustness for sample instances. For example, a sample instance can be
correctly classified by a model trained using standard adversarial training
(AT) but not by a model trained using TRADES, and vice versa. Based on this
observation, we propose a collaborative adversarial training framework to
improve the robustness of neural networks. Specifically, we use different
adversarial training methods to train robust models and let models interact
with their knowledge during the training process. Collaborative Adversarial
Training (CAT) can improve both robustness and accuracy. Extensive experiments
on various networks and datasets validate the effectiveness of our method. CAT
achieves state-of-the-art adversarial robustness without using any additional
data on CIFAR-10 under the Auto-Attack benchmark. Code is available at
https://github.com/liuxingbin/CAT.
| no_new_dataset | 0.950273 |
2303.14926 | Kun Hu | Clinton Ansun Mo, Kun Hu, Chengjiang Long, Zhiyong Wang | Continuous Intermediate Token Learning with Implicit Motion Manifold for
Keyframe Based Motion Interpolation | Accepted by CVPR 2023 | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by/4.0/ | Deriving sophisticated 3D motions from sparse keyframes is a particularly
challenging problem, due to continuity and exceptionally skeletal precision.
The action features are often derivable accurately from the full series of
keyframes, and thus, leveraging the global context with transformers has been a
promising data-driven embedding approach. However, existing methods are often
with inputs of interpolated intermediate frame for continuity using basic
interpolation methods with keyframes, which result in a trivial local minimum
during training. In this paper, we propose a novel framework to formulate
latent motion manifolds with keyframe-based constraints, from which the
continuous nature of intermediate token representations is considered.
Particularly, our proposed framework consists of two stages for identifying a
latent motion subspace, i.e., a keyframe encoding stage and an intermediate
token generation stage, and a subsequent motion synthesis stage to extrapolate
and compose motion data from manifolds. Through our extensive experiments
conducted on both the LaFAN1 and CMU Mocap datasets, our proposed method
demonstrates both superior interpolation accuracy and high visual similarity to
ground truth motions.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 05:53:01 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Mo",
"Clinton Ansun",
""
],
[
"Hu",
"Kun",
""
],
[
"Long",
"Chengjiang",
""
],
[
"Wang",
"Zhiyong",
""
]
] | TITLE: Continuous Intermediate Token Learning with Implicit Motion Manifold for
Keyframe Based Motion Interpolation
ABSTRACT: Deriving sophisticated 3D motions from sparse keyframes is a particularly
challenging problem, due to continuity and exceptionally skeletal precision.
The action features are often derivable accurately from the full series of
keyframes, and thus, leveraging the global context with transformers has been a
promising data-driven embedding approach. However, existing methods are often
with inputs of interpolated intermediate frame for continuity using basic
interpolation methods with keyframes, which result in a trivial local minimum
during training. In this paper, we propose a novel framework to formulate
latent motion manifolds with keyframe-based constraints, from which the
continuous nature of intermediate token representations is considered.
Particularly, our proposed framework consists of two stages for identifying a
latent motion subspace, i.e., a keyframe encoding stage and an intermediate
token generation stage, and a subsequent motion synthesis stage to extrapolate
and compose motion data from manifolds. Through our extensive experiments
conducted on both the LaFAN1 and CMU Mocap datasets, our proposed method
demonstrates both superior interpolation accuracy and high visual similarity to
ground truth motions.
| no_new_dataset | 0.942348 |
2303.14935 | Nam Ly Tuan | Phuc Nguyen, Nam Tuan Ly, Hideaki Takeda, and Atsuhiro Takasu | TabIQA: Table Questions Answering on Business Document Images | First two authors contributed equally | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Table answering questions from business documents has many challenges that
require understanding tabular structures, cross-document referencing, and
additional numeric computations beyond simple search queries. This paper
introduces a novel pipeline, named TabIQA, to answer questions about business
document images. TabIQA combines state-of-the-art deep learning techniques 1)
to extract table content and structural information from images and 2) to
answer various questions related to numerical data, text-based information, and
complex queries from structured tables. The evaluation results on VQAonBD 2023
dataset demonstrate the effectiveness of TabIQA in achieving promising
performance in answering table-related questions. The TabIQA repository is
available at https://github.com/phucty/itabqa.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 06:31:21 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Nguyen",
"Phuc",
""
],
[
"Ly",
"Nam Tuan",
""
],
[
"Takeda",
"Hideaki",
""
],
[
"Takasu",
"Atsuhiro",
""
]
] | TITLE: TabIQA: Table Questions Answering on Business Document Images
ABSTRACT: Table answering questions from business documents has many challenges that
require understanding tabular structures, cross-document referencing, and
additional numeric computations beyond simple search queries. This paper
introduces a novel pipeline, named TabIQA, to answer questions about business
document images. TabIQA combines state-of-the-art deep learning techniques 1)
to extract table content and structural information from images and 2) to
answer various questions related to numerical data, text-based information, and
complex queries from structured tables. The evaluation results on VQAonBD 2023
dataset demonstrate the effectiveness of TabIQA in achieving promising
performance in answering table-related questions. The TabIQA repository is
available at https://github.com/phucty/itabqa.
| no_new_dataset | 0.926636 |
2303.14937 | \c{C}a\u{g}lar Aytekin | Caglar Aytekin | LEURN: Learning Explainable Univariate Rules with Neural Networks | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In this paper, we propose LEURN: a neural network architecture that learns
univariate decision rules. LEURN is a white-box algorithm that results into
univariate trees and makes explainable decisions in every stage. In each layer,
LEURN finds a set of univariate rules based on an embedding of the previously
checked rules and their corresponding responses. Both rule finding and final
decision mechanisms are weighted linear combinations of these embeddings, hence
contribution of all rules are clearly formulated and explainable. LEURN can
select features, extract feature importance, provide semantic similarity
between a pair of samples, be used in a generative manner and can give a
confidence score. Thanks to a smoothness parameter, LEURN can also controllably
behave like decision trees or vanilla neural networks. Besides these
advantages, LEURN achieves comparable performance to state-of-the-art methods
across 30 tabular datasets for classification and regression problems.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 06:34:42 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Aytekin",
"Caglar",
""
]
] | TITLE: LEURN: Learning Explainable Univariate Rules with Neural Networks
ABSTRACT: In this paper, we propose LEURN: a neural network architecture that learns
univariate decision rules. LEURN is a white-box algorithm that results into
univariate trees and makes explainable decisions in every stage. In each layer,
LEURN finds a set of univariate rules based on an embedding of the previously
checked rules and their corresponding responses. Both rule finding and final
decision mechanisms are weighted linear combinations of these embeddings, hence
contribution of all rules are clearly formulated and explainable. LEURN can
select features, extract feature importance, provide semantic similarity
between a pair of samples, be used in a generative manner and can give a
confidence score. Thanks to a smoothness parameter, LEURN can also controllably
behave like decision trees or vanilla neural networks. Besides these
advantages, LEURN achieves comparable performance to state-of-the-art methods
across 30 tabular datasets for classification and regression problems.
| no_new_dataset | 0.946547 |
2303.14939 | Williams Rizzi | Williams Rizzi and Chiara Di Francescomarino and Chiara Ghidini and
Fabrizio Maria Maggi | Explain, Adapt and Retrain: How to improve the accuracy of a PPM
classifier through different explanation styles | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent papers have introduced a novel approach to explain why a Predictive
Process Monitoring (PPM) model for outcome-oriented predictions provides wrong
predictions. Moreover, they have shown how to exploit the explanations,
obtained using state-of-the art post-hoc explainers, to identify the most
common features that induce a predictor to make mistakes in a semi-automated
way, and, in turn, to reduce the impact of those features and increase the
accuracy of the predictive model. This work starts from the assumption that
frequent control flow patterns in event logs may represent important features
that characterize, and therefore explain, a certain prediction. Therefore, in
this paper, we (i) employ a novel encoding able to leverage DECLARE constraints
in Predictive Process Monitoring and compare the effectiveness of this encoding
with Predictive Process Monitoring state-of-the art encodings, in particular
for the task of outcome-oriented predictions; (ii) introduce a completely
automated pipeline for the identification of the most common features inducing
a predictor to make mistakes; and (iii) show the effectiveness of the proposed
pipeline in increasing the accuracy of the predictive model by validating it on
different real-life datasets.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 06:37:55 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Rizzi",
"Williams",
""
],
[
"Di Francescomarino",
"Chiara",
""
],
[
"Ghidini",
"Chiara",
""
],
[
"Maggi",
"Fabrizio Maria",
""
]
] | TITLE: Explain, Adapt and Retrain: How to improve the accuracy of a PPM
classifier through different explanation styles
ABSTRACT: Recent papers have introduced a novel approach to explain why a Predictive
Process Monitoring (PPM) model for outcome-oriented predictions provides wrong
predictions. Moreover, they have shown how to exploit the explanations,
obtained using state-of-the art post-hoc explainers, to identify the most
common features that induce a predictor to make mistakes in a semi-automated
way, and, in turn, to reduce the impact of those features and increase the
accuracy of the predictive model. This work starts from the assumption that
frequent control flow patterns in event logs may represent important features
that characterize, and therefore explain, a certain prediction. Therefore, in
this paper, we (i) employ a novel encoding able to leverage DECLARE constraints
in Predictive Process Monitoring and compare the effectiveness of this encoding
with Predictive Process Monitoring state-of-the art encodings, in particular
for the task of outcome-oriented predictions; (ii) introduce a completely
automated pipeline for the identification of the most common features inducing
a predictor to make mistakes; and (iii) show the effectiveness of the proposed
pipeline in increasing the accuracy of the predictive model by validating it on
different real-life datasets.
| no_new_dataset | 0.944074 |
2303.14951 | Suman Adhya | Suman Adhya, Avishek Lahiri, Debarshi Kumar Sanyal, Partha Pratim Das | Improving Contextualized Topic Models with Negative Sampling | Accepted at 19th International Conference on Natural Language
Processing (ICON 2022) | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Topic modeling has emerged as a dominant method for exploring large document
collections. Recent approaches to topic modeling use large contextualized
language models and variational autoencoders. In this paper, we propose a
negative sampling mechanism for a contextualized topic model to improve the
quality of the generated topics. In particular, during model training, we
perturb the generated document-topic vector and use a triplet loss to encourage
the document reconstructed from the correct document-topic vector to be similar
to the input document and dissimilar to the document reconstructed from the
perturbed vector. Experiments for different topic counts on three publicly
available benchmark datasets show that in most cases, our approach leads to an
increase in topic coherence over that of the baselines. Our model also achieves
very high topic diversity.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 07:28:46 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Adhya",
"Suman",
""
],
[
"Lahiri",
"Avishek",
""
],
[
"Sanyal",
"Debarshi Kumar",
""
],
[
"Das",
"Partha Pratim",
""
]
] | TITLE: Improving Contextualized Topic Models with Negative Sampling
ABSTRACT: Topic modeling has emerged as a dominant method for exploring large document
collections. Recent approaches to topic modeling use large contextualized
language models and variational autoencoders. In this paper, we propose a
negative sampling mechanism for a contextualized topic model to improve the
quality of the generated topics. In particular, during model training, we
perturb the generated document-topic vector and use a triplet loss to encourage
the document reconstructed from the correct document-topic vector to be similar
to the input document and dissimilar to the document reconstructed from the
perturbed vector. Experiments for different topic counts on three publicly
available benchmark datasets show that in most cases, our approach leads to an
increase in topic coherence over that of the baselines. Our model also achieves
very high topic diversity.
| no_new_dataset | 0.952086 |
2303.14953 | Ming Wang | Ming Wang, Xianda Guo, Beibei Lin, Tian Yang, Zheng Zhu, Lincheng Li,
Shunli Zhang and Xin Yu | DyGait: Exploiting Dynamic Representations for High-performance Gait
Recognition | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Gait recognition is a biometric technology that recognizes the identity of
humans through their walking patterns. Compared with other biometric
technologies, gait recognition is more difficult to disguise and can be applied
to the condition of long-distance without the cooperation of subjects. Thus, it
has unique potential and wide application for crime prevention and social
security. At present, most gait recognition methods directly extract features
from the video frames to establish representations. However, these
architectures learn representations from different features equally but do not
pay enough attention to dynamic features, which refers to a representation of
dynamic parts of silhouettes over time (e.g. legs). Since dynamic parts of the
human body are more informative than other parts (e.g. bags) during walking, in
this paper, we propose a novel and high-performance framework named DyGait.
This is the first framework on gait recognition that is designed to focus on
the extraction of dynamic features. Specifically, to take full advantage of the
dynamic information, we propose a Dynamic Augmentation Module (DAM), which can
automatically establish spatial-temporal feature representations of the dynamic
parts of the human body. The experimental results show that our DyGait network
outperforms other state-of-the-art gait recognition methods. It achieves an
average Rank-1 accuracy of 71.4% on the GREW dataset, 66.3% on the Gait3D
dataset, 98.4% on the CASIA-B dataset and 98.3% on the OU-MVLP dataset.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 07:36:47 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Wang",
"Ming",
""
],
[
"Guo",
"Xianda",
""
],
[
"Lin",
"Beibei",
""
],
[
"Yang",
"Tian",
""
],
[
"Zhu",
"Zheng",
""
],
[
"Li",
"Lincheng",
""
],
[
"Zhang",
"Shunli",
""
],
[
"Yu",
"Xin",
""
]
] | TITLE: DyGait: Exploiting Dynamic Representations for High-performance Gait
Recognition
ABSTRACT: Gait recognition is a biometric technology that recognizes the identity of
humans through their walking patterns. Compared with other biometric
technologies, gait recognition is more difficult to disguise and can be applied
to the condition of long-distance without the cooperation of subjects. Thus, it
has unique potential and wide application for crime prevention and social
security. At present, most gait recognition methods directly extract features
from the video frames to establish representations. However, these
architectures learn representations from different features equally but do not
pay enough attention to dynamic features, which refers to a representation of
dynamic parts of silhouettes over time (e.g. legs). Since dynamic parts of the
human body are more informative than other parts (e.g. bags) during walking, in
this paper, we propose a novel and high-performance framework named DyGait.
This is the first framework on gait recognition that is designed to focus on
the extraction of dynamic features. Specifically, to take full advantage of the
dynamic information, we propose a Dynamic Augmentation Module (DAM), which can
automatically establish spatial-temporal feature representations of the dynamic
parts of the human body. The experimental results show that our DyGait network
outperforms other state-of-the-art gait recognition methods. It achieves an
average Rank-1 accuracy of 71.4% on the GREW dataset, 66.3% on the Gait3D
dataset, 98.4% on the CASIA-B dataset and 98.3% on the OU-MVLP dataset.
| no_new_dataset | 0.939304 |
2303.14964 | Haoyu Chen | Haoyu Chen, Zhihua Wang, Yang Yang, Qilin Sun, Kede Ma | Learning a Deep Color Difference Metric for Photographic Images | null | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most well-established and widely used color difference (CD) metrics are
handcrafted and subject-calibrated against uniformly colored patches, which do
not generalize well to photographic images characterized by natural scene
complexities. Constructing CD formulae for photographic images is still an
active research topic in imaging/illumination, vision science, and color
science communities. In this paper, we aim to learn a deep CD metric for
photographic images with four desirable properties. First, it well aligns with
the observations in vision science that color and form are linked inextricably
in visual cortical processing. Second, it is a proper metric in the
mathematical sense. Third, it computes accurate CDs between photographic
images, differing mainly in color appearances. Fourth, it is robust to mild
geometric distortions (e.g., translation or due to parallax), which are often
present in photographic images of the same scene captured by different digital
cameras. We show that all these properties can be satisfied at once by learning
a multi-scale autoregressive normalizing flow for feature transform, followed
by the Euclidean distance which is linearly proportional to the human
perceptual CD. Quantitative and qualitative experiments on the large-scale SPCD
dataset demonstrate the promise of the learned CD metric.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 07:54:09 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Chen",
"Haoyu",
""
],
[
"Wang",
"Zhihua",
""
],
[
"Yang",
"Yang",
""
],
[
"Sun",
"Qilin",
""
],
[
"Ma",
"Kede",
""
]
] | TITLE: Learning a Deep Color Difference Metric for Photographic Images
ABSTRACT: Most well-established and widely used color difference (CD) metrics are
handcrafted and subject-calibrated against uniformly colored patches, which do
not generalize well to photographic images characterized by natural scene
complexities. Constructing CD formulae for photographic images is still an
active research topic in imaging/illumination, vision science, and color
science communities. In this paper, we aim to learn a deep CD metric for
photographic images with four desirable properties. First, it well aligns with
the observations in vision science that color and form are linked inextricably
in visual cortical processing. Second, it is a proper metric in the
mathematical sense. Third, it computes accurate CDs between photographic
images, differing mainly in color appearances. Fourth, it is robust to mild
geometric distortions (e.g., translation or due to parallax), which are often
present in photographic images of the same scene captured by different digital
cameras. We show that all these properties can be satisfied at once by learning
a multi-scale autoregressive normalizing flow for feature transform, followed
by the Euclidean distance which is linearly proportional to the human
perceptual CD. Quantitative and qualitative experiments on the large-scale SPCD
dataset demonstrate the promise of the learned CD metric.
| no_new_dataset | 0.950503 |
2303.14968 | Weixia Zhang | Weixia Zhang and Guangtao Zhai and Ying Wei and Xiaokang Yang and Kede
Ma | Blind Image Quality Assessment via Vision-Language Correspondence: A
Multitask Learning Perspective | CVPR2023 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | We aim at advancing blind image quality assessment (BIQA), which predicts the
human perception of image quality without any reference information. We develop
a general and automated multitask learning scheme for BIQA to exploit auxiliary
knowledge from other tasks, in a way that the model parameter sharing and the
loss weighting are determined automatically. Specifically, we first describe
all candidate label combinations (from multiple tasks) using a textual
template, and compute the joint probability from the cosine similarities of the
visual-textual embeddings. Predictions of each task can be inferred from the
joint distribution, and optimized by carefully designed loss functions. Through
comprehensive experiments on learning three tasks - BIQA, scene classification,
and distortion type identification, we verify that the proposed BIQA method 1)
benefits from the scene classification and distortion type identification tasks
and outperforms the state-of-the-art on multiple IQA datasets, 2) is more
robust in the group maximum differentiation competition, and 3) realigns the
quality annotations from different IQA datasets more effectively. The source
code is available at https://github.com/zwx8981/LIQE.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 07:58:09 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Zhang",
"Weixia",
""
],
[
"Zhai",
"Guangtao",
""
],
[
"Wei",
"Ying",
""
],
[
"Yang",
"Xiaokang",
""
],
[
"Ma",
"Kede",
""
]
] | TITLE: Blind Image Quality Assessment via Vision-Language Correspondence: A
Multitask Learning Perspective
ABSTRACT: We aim at advancing blind image quality assessment (BIQA), which predicts the
human perception of image quality without any reference information. We develop
a general and automated multitask learning scheme for BIQA to exploit auxiliary
knowledge from other tasks, in a way that the model parameter sharing and the
loss weighting are determined automatically. Specifically, we first describe
all candidate label combinations (from multiple tasks) using a textual
template, and compute the joint probability from the cosine similarities of the
visual-textual embeddings. Predictions of each task can be inferred from the
joint distribution, and optimized by carefully designed loss functions. Through
comprehensive experiments on learning three tasks - BIQA, scene classification,
and distortion type identification, we verify that the proposed BIQA method 1)
benefits from the scene classification and distortion type identification tasks
and outperforms the state-of-the-art on multiple IQA datasets, 2) is more
robust in the group maximum differentiation competition, and 3) realigns the
quality annotations from different IQA datasets more effectively. The source
code is available at https://github.com/zwx8981/LIQE.
| no_new_dataset | 0.944536 |
2303.14969 | Donggyun Kim | Donggyun Kim, Jinwoo Kim, Seongwoong Cho, Chong Luo, Seunghoon Hong | Universal Few-shot Learning of Dense Prediction Tasks with Visual Token
Matching | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Dense prediction tasks are a fundamental class of problems in computer
vision. As supervised methods suffer from high pixel-wise labeling cost, a
few-shot learning solution that can learn any dense task from a few labeled
images is desired. Yet, current few-shot learning methods target a restricted
set of tasks such as semantic segmentation, presumably due to challenges in
designing a general and unified model that is able to flexibly and efficiently
adapt to arbitrary tasks of unseen semantics. We propose Visual Token Matching
(VTM), a universal few-shot learner for arbitrary dense prediction tasks. It
employs non-parametric matching on patch-level embedded tokens of images and
labels that encapsulates all tasks. Also, VTM flexibly adapts to any task with
a tiny amount of task-specific parameters that modulate the matching algorithm.
We implement VTM as a powerful hierarchical encoder-decoder architecture
involving ViT backbones where token matching is performed at multiple feature
hierarchies. We experiment VTM on a challenging variant of Taskonomy dataset
and observe that it robustly few-shot learns various unseen dense prediction
tasks. Surprisingly, it is competitive with fully supervised baselines using
only 10 labeled examples of novel tasks (0.004% of full supervision) and
sometimes outperforms using 0.1% of full supervision. Codes are available at
https://github.com/GitGyun/visual_token_matching.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 07:58:42 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Kim",
"Donggyun",
""
],
[
"Kim",
"Jinwoo",
""
],
[
"Cho",
"Seongwoong",
""
],
[
"Luo",
"Chong",
""
],
[
"Hong",
"Seunghoon",
""
]
] | TITLE: Universal Few-shot Learning of Dense Prediction Tasks with Visual Token
Matching
ABSTRACT: Dense prediction tasks are a fundamental class of problems in computer
vision. As supervised methods suffer from high pixel-wise labeling cost, a
few-shot learning solution that can learn any dense task from a few labeled
images is desired. Yet, current few-shot learning methods target a restricted
set of tasks such as semantic segmentation, presumably due to challenges in
designing a general and unified model that is able to flexibly and efficiently
adapt to arbitrary tasks of unseen semantics. We propose Visual Token Matching
(VTM), a universal few-shot learner for arbitrary dense prediction tasks. It
employs non-parametric matching on patch-level embedded tokens of images and
labels that encapsulates all tasks. Also, VTM flexibly adapts to any task with
a tiny amount of task-specific parameters that modulate the matching algorithm.
We implement VTM as a powerful hierarchical encoder-decoder architecture
involving ViT backbones where token matching is performed at multiple feature
hierarchies. We experiment VTM on a challenging variant of Taskonomy dataset
and observe that it robustly few-shot learns various unseen dense prediction
tasks. Surprisingly, it is competitive with fully supervised baselines using
only 10 labeled examples of novel tasks (0.004% of full supervision) and
sometimes outperforms using 0.1% of full supervision. Codes are available at
https://github.com/GitGyun/visual_token_matching.
| no_new_dataset | 0.945248 |
2303.14977 | Xiaohui Guo | Xiaohui Guo | A novel Multi to Single Module for small object detection | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Small object detection presents a significant challenge in computer vision
and object detection. The performance of small object detectors is often
compromised by a lack of pixels and less significant features. This issue stems
from information misalignment caused by variations in feature scale and
information loss during feature processing. In response to this challenge, this
paper proposes a novel the Multi to Single Module (M2S), which enhances a
specific layer through improving feature extraction and refining features.
Specifically, M2S includes the proposed Cross-scale Aggregation Module (CAM)
and explored Dual Relationship Module (DRM) to improve information extraction
capabilities and feature refinement effects. Moreover, this paper enhances the
accuracy of small object detection by utilizing M2S to generate an additional
detection head. The effectiveness of the proposed method is evaluated on two
datasets, VisDrone2021-DET and SeaDronesSeeV2. The experimental results
demonstrate its improved performance compared with existing methods. Compared
to the baseline model (YOLOv5s), M2S improves the accuracy by about 1.1\% on
the VisDrone2021-DET testing dataset and 15.68\% on the SeaDronesSeeV2
validation set.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 08:17:22 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Guo",
"Xiaohui",
""
]
] | TITLE: A novel Multi to Single Module for small object detection
ABSTRACT: Small object detection presents a significant challenge in computer vision
and object detection. The performance of small object detectors is often
compromised by a lack of pixels and less significant features. This issue stems
from information misalignment caused by variations in feature scale and
information loss during feature processing. In response to this challenge, this
paper proposes a novel the Multi to Single Module (M2S), which enhances a
specific layer through improving feature extraction and refining features.
Specifically, M2S includes the proposed Cross-scale Aggregation Module (CAM)
and explored Dual Relationship Module (DRM) to improve information extraction
capabilities and feature refinement effects. Moreover, this paper enhances the
accuracy of small object detection by utilizing M2S to generate an additional
detection head. The effectiveness of the proposed method is evaluated on two
datasets, VisDrone2021-DET and SeaDronesSeeV2. The experimental results
demonstrate its improved performance compared with existing methods. Compared
to the baseline model (YOLOv5s), M2S improves the accuracy by about 1.1\% on
the VisDrone2021-DET testing dataset and 15.68\% on the SeaDronesSeeV2
validation set.
| no_new_dataset | 0.950686 |
2303.14978 | Jinming Liu | Jinming Liu, Heming Sun, Jiro Katto | Learned Image Compression with Mixed Transformer-CNN Architectures | Accepted by CVPR2023 (Highlight) | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learned image compression (LIC) methods have exhibited promising progress and
superior rate-distortion performance compared with classical image compression
standards. Most existing LIC methods are Convolutional Neural Networks-based
(CNN-based) or Transformer-based, which have different advantages. Exploiting
both advantages is a point worth exploring, which has two challenges: 1) how to
effectively fuse the two methods? 2) how to achieve higher performance with a
suitable complexity? In this paper, we propose an efficient parallel
Transformer-CNN Mixture (TCM) block with a controllable complexity to
incorporate the local modeling ability of CNN and the non-local modeling
ability of transformers to improve the overall architecture of image
compression models. Besides, inspired by the recent progress of entropy
estimation models and attention modules, we propose a channel-wise entropy
model with parameter-efficient swin-transformer-based attention (SWAtten)
modules by using channel squeezing. Experimental results demonstrate our
proposed method achieves state-of-the-art rate-distortion performances on three
different resolution datasets (i.e., Kodak, Tecnick, CLIC Professional
Validation) compared to existing LIC methods. The code is at
https://github.com/jmliu206/LIC_TCM.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 08:19:01 GMT"
}
] | 2023-03-28T00:00:00 | [
[
"Liu",
"Jinming",
""
],
[
"Sun",
"Heming",
""
],
[
"Katto",
"Jiro",
""
]
] | TITLE: Learned Image Compression with Mixed Transformer-CNN Architectures
ABSTRACT: Learned image compression (LIC) methods have exhibited promising progress and
superior rate-distortion performance compared with classical image compression
standards. Most existing LIC methods are Convolutional Neural Networks-based
(CNN-based) or Transformer-based, which have different advantages. Exploiting
both advantages is a point worth exploring, which has two challenges: 1) how to
effectively fuse the two methods? 2) how to achieve higher performance with a
suitable complexity? In this paper, we propose an efficient parallel
Transformer-CNN Mixture (TCM) block with a controllable complexity to
incorporate the local modeling ability of CNN and the non-local modeling
ability of transformers to improve the overall architecture of image
compression models. Besides, inspired by the recent progress of entropy
estimation models and attention modules, we propose a channel-wise entropy
model with parameter-efficient swin-transformer-based attention (SWAtten)
modules by using channel squeezing. Experimental results demonstrate our
proposed method achieves state-of-the-art rate-distortion performances on three
different resolution datasets (i.e., Kodak, Tecnick, CLIC Professional
Validation) compared to existing LIC methods. The code is at
https://github.com/jmliu206/LIC_TCM.
| no_new_dataset | 0.946646 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.