parent_paper_title
stringclasses
63 values
parent_paper_arxiv_id
stringclasses
63 values
citation_shorthand
stringlengths
2
56
raw_citation_text
stringlengths
9
63
cited_paper_title
stringlengths
5
161
cited_paper_arxiv_link
stringlengths
32
37
cited_paper_abstract
stringlengths
406
1.92k
has_metadata
bool
1 class
is_arxiv_paper
bool
2 classes
bib_paper_authors
stringlengths
2
2.44k
bib_paper_year
float64
1.97k
2.03k
bib_paper_month
stringclasses
16 values
bib_paper_url
stringlengths
20
116
bib_paper_doi
stringclasses
269 values
bib_paper_journal
stringlengths
3
148
original_title
stringlengths
5
161
search_res_title
stringlengths
4
122
search_res_url
stringlengths
22
267
search_res_content
stringlengths
19
1.92k
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
fang2018learning
\cite{fang2018learning}
Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation
http://arxiv.org/abs/1710.06513v6
In this paper, we propose a pose grammar to tackle the problem of 3D human pose estimation. Our model directly takes 2D pose as input and learns a generalized 2D-3D mapping function. The proposed model consists of a base network which efficiently captures pose-aligned features and a hierarchy of Bi-directional RNNs (BR...
true
true
Fang, Hao-Shu and Xu, Yuanlu and Wang, Wenguan and Liu, Xiaobai and Zhu, Song-Chun
2,018
null
null
null
null
Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation
[PDF] Learning Pose Grammar to Encode Human Body Configuration for ...
https://cdn.aaai.org/ojs/12270/12270-13-15798-1-2-20201228.pdf
In this paper, we propose a pose grammar to tackle the prob- lem of 3D human pose estimation. Our model directly takes. 2D pose as input and learns a
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
he2021db
\cite{he2021db}
{DB-LSTM: Densely-connected Bi-directional LSTM for human action recognition}
null
null
true
false
He, Jun-Yan and Wu, Xiao and Cheng, Zhi-Qi and Yuan, Zhaoquan and Jiang, Yu-Gang
2,021
null
null
null
Neurocomputing
{DB-LSTM: Densely-connected Bi-directional LSTM for human action recognition}
Densely-connected Bi-directional LSTM for human action ...
https://www.sciencedirect.com/science/article/pii/S0925231220317859
To boost the effectiveness and robustness of modeling long-range action recognition, a Densely-connected Bi-directional LSTM (DB-LSTM) network is novelly proposed to model the visual and temporal associations in both forward and backward directions. To overcome the drawbacks of existing methods, a long-range temporal m...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
zeng2021learning
\cite{zeng2021learning}
Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation
http://arxiv.org/abs/2108.07181v2
Various deep learning techniques have been proposed to solve the single-view 2D-to-3D pose estimation problem. While the average prediction accuracy has been improved significantly over the years, the performance on hard poses with depth ambiguity, self-occlusion, and complex or rare poses is still far from satisfactor...
true
true
Zeng, Ailing and Sun, Xiao and Yang, Lei and Zhao, Nanxuan and Liu, Minhao and Xu, Qiang
2,021
null
null
null
null
Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation
Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation
http://arxiv.org/pdf/2108.07181v2
Various deep learning techniques have been proposed to solve the single-view 2D-to-3D pose estimation problem. While the average prediction accuracy has been improved significantly over the years, the performance on hard poses with depth ambiguity, self-occlusion, and complex or rare poses is still far from satisfactor...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
zhang2023learning
\cite{zhang2023learning}
Learning Enriched Hop-Aware Correlation for Robust {3D} Human Pose Estimation
null
null
true
false
Zhang, Shengping and Wang, Chenyang and Nie, Liqiang and Yao, Hongxun and Huang, Qingming and Tian, Qi
2,023
null
null
null
International Journal of Computer Vision
Learning Enriched Hop-Aware Correlation for Robust {3D} Human Pose Estimation
Learning Enriched Hop-Aware Correlation for Robust 3D Human ...
https://link.springer.com/article/10.1007/s11263-023-01770-5
This paper proposes a parallel hop-aware graph attention network (PHGANet) for 3D human pose estimation, which learns enriched hop-aware correlation of the
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
li2022exploiting
\cite{li2022exploiting}
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
http://arxiv.org/abs/2103.14304v8
Despite the great progress in 3D human pose estimation from videos, it is still an open problem to take full advantage of a redundant 2D pose sequence to learn representative representations for generating one 3D pose. To this end, we propose an improved Transformer-based architecture, called Strided Transformer, which...
true
true
Li, Wenhao and Liu, Hong and Ding, Runwei and Liu, Mengyuan and Wang, Pichao and Yang, Wenming
2,022
null
null
null
IEEE Transactions on Multimedia
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation
Vegetebird/StridedTransformer-Pose3D
https://github.com/Vegetebird/StridedTransformer-Pose3D
Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimation. This is the official implementation of the approach described in the paper.
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
zhang2022mixste
\cite{zhang2022mixste}
MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video
http://arxiv.org/abs/2203.00859v4
Recent transformer-based solutions have been introduced to estimate 3D human pose from 2D keypoint sequence by considering body joints among all frames globally to learn spatio-temporal correlation. We observe that the motions of different joints differ significantly. However, the previous methods cannot efficiently mo...
true
true
Zhang, Jinlu and Tu, Zhigang and Yang, Jianyu and Chen, Yujin and Yuan, Junsong
2,022
null
null
null
null
MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video
MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human ...
https://github.com/JinluZhang1126/MixSTE
Official implementation of CVPR 2022 paper(MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video).
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
vaswani2017attention
\cite{vaswani2017attention}
Attention Is All You Need
http://arxiv.org/abs/1706.03762v7
The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on at...
true
true
Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N and Kaiser, {\L}ukasz and Polosukhin, Illia
2,017
null
null
null
Advances in Neural Information Processing Systems
Attention Is All You Need
Attention Is All You Need
http://arxiv.org/pdf/1706.03762v7
The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on at...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
zhou2019hemlets
\cite{zhou2019hemlets}
HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation
http://arxiv.org/abs/1910.12032v1
Estimating 3D human pose from a single image is a challenging task. This work attempts to address the uncertainty of lifting the detected 2D joints to the 3D space by introducing an intermediate state - Part-Centric Heatmap Triplets (HEMlets), which shortens the gap between the 2D observation and the 3D interpretation....
true
true
Zhou, Kun and Han, Xiaoguang and Jiang, Nianjuan and Jia, Kui and Lu, Jiangbo
2,019
null
null
null
null
HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation
redrock303/HEMlets
https://github.com/redrock303/HEMlets
Here we provide our implementation of HEMlets PoSh: Learning Part-Centric Heatmap Triplets for 3D Human Pose and Shape Estimation.
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
zeng2020srnet
\cite{zeng2020srnet}
SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach
http://arxiv.org/abs/2007.09389v1
Human poses that are rare or unseen in a training set are challenging for a network to predict. Similar to the long-tailed distribution problem in visual recognition, the small number of examples for such poses limits the ability of networks to model them. Interestingly, local pose distributions suffer less from the lo...
true
true
Zeng, Ailing and Sun, Xiao and Huang, Fuyang and Liu, Minhao and Xu, Qiang and Lin, Stephen
2,020
null
null
null
null
SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach
GitHub - ailingzengzzz/Split-and-Recombine-Net: Code for "SRNet
https://github.com/ailingzengzzz/Split-and-Recombine-Net
This is the original PyTorch implementation of the following work: SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
xue2022boosting
\cite{xue2022boosting}
Boosting monocular {3D} human pose estimation with part aware attention
null
null
true
false
Xue, Youze and Chen, Jiansheng and Gu, Xiangming and Ma, Huimin and Ma, Hongbing
2,022
null
null
null
IEEE Transactions on Image Processing
Boosting monocular {3D} human pose estimation with part aware attention
Boosting Monocular 3D Human Pose Estimation With Part Aware ...
https://ieeexplore.ieee.org/iel7/83/9626658/09798770.pdf
We thus propose the Part Aware. Dictionary Attention module to calculate the attention for the part-wise features of input in a dictionary, which contains.
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
wu2022hpgcn
\cite{wu2022hpgcn}
{HPGCN: Hierarchical poselet-guided graph convolutional network for {3D} pose estimation}
null
null
true
false
Wu, Yongpeng and Kong, Dehui and Wang, Shaofan and Li, Jinghua and Yin, Baocai
2,022
null
null
null
Neurocomputing
{HPGCN: Hierarchical poselet-guided graph convolutional network for {3D} pose estimation}
HPGCN: Hierarchical poselet-guided graph convolutional network ...
https://www.sciencedirect.com/science/article/pii/S0925231221016817
We propose a hierarchical poselet-guided graph convolutional network (HPGCN) for 3D pose estimation from 2D poses.
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
xu2021graph
\cite{xu2021graph}
Graph Stacked Hourglass Networks for 3D Human Pose Estimation
http://arxiv.org/abs/2103.16385v1
In this paper, we propose a novel graph convolutional network architecture, Graph Stacked Hourglass Networks, for 2D-to-3D human pose estimation tasks. The proposed architecture consists of repeated encoder-decoder, in which graph-structured features are processed across three different scales of human skeletal represe...
true
true
Xu, Tianhan and Takano, Wataru
2,021
null
null
null
null
Graph Stacked Hourglass Networks for 3D Human Pose Estimation
Graph Stacked Hourglass Networks for 3D Human Pose Estimation
http://arxiv.org/pdf/2103.16385v1
In this paper, we propose a novel graph convolutional network architecture, Graph Stacked Hourglass Networks, for 2D-to-3D human pose estimation tasks. The proposed architecture consists of repeated encoder-decoder, in which graph-structured features are processed across three different scales of human skeletal represe...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
hua2022unet
\cite{hua2022unet}
Weakly-supervised {3D} human pose estimation with cross-view U-shaped graph convolutional network
null
null
true
false
Hua, Guoliang and Liu, Hong and Li, Wenhao and Zhang, Qian and Ding, Runwei and Xu, Xin
2,022
null
null
null
IEEE Transactions on Multimedia
Weakly-supervised {3D} human pose estimation with cross-view U-shaped graph convolutional network
Weakly-supervised 3D Human Pose Estimation with Cross-view U ...
https://arxiv.org/abs/2105.10882
[2105.10882] Weakly-supervised 3D Human Pose Estimation with Cross-view U-shaped Graph Convolutional Network **arXiv:2105.10882** (cs) Title:Weakly-supervised 3D Human Pose Estimation with Cross-view U-shaped Graph Convolutional Network View a PDF of the paper titled Weakly-supervised 3D Human Pose Estimation with Cro...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
wu2022p2t
\cite{wu2022p2t}
P2T: Pyramid Pooling Transformer for Scene Understanding
http://arxiv.org/abs/2106.12011v6
Recently, the vision transformer has achieved great success by pushing the state-of-the-art of various vision tasks. One of the most challenging problems in the vision transformer is that the large sequence length of image tokens leads to high computational cost (quadratic complexity). A popular solution to this proble...
true
true
Wu, Yu-Huan and Liu, Yun and Zhan, Xin and Cheng, Ming-Ming
2,022
null
null
null
IEEE Transactions on Pattern Analysis and Machine Intelligence
P2T: Pyramid Pooling Transformer for Scene Understanding
P2T: Pyramid Pooling Transformer for Scene Understanding
http://arxiv.org/pdf/2106.12011v6
Recently, the vision transformer has achieved great success by pushing the state-of-the-art of various vision tasks. One of the most challenging problems in the vision transformer is that the large sequence length of image tokens leads to high computational cost (quadratic complexity). A popular solution to this proble...
Learning Pyramid-structured Long-range Dependencies for 3D Human Pose Estimation
2506.02853v1
PVT
\cite{PVT}
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
http://arxiv.org/abs/2102.12122v2
Although using convolutional neural networks (CNNs) as backbones achieves great successes in computer vision, this work investigates a simple backbone network useful for many dense prediction tasks without convolutions. Unlike the recently-proposed Transformer model (e.g., ViT) that is specially designed for image clas...
true
true
Wang, Wenhai and Xie, Enze and Li, Xiang and Fan, Deng-Ping and Song, Kaitao and Liang, Ding and Lu, Tong and Luo, Ping and Shao, Ling
2,021
null
null
null
null
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions
Pyramid Vision Transformer: A Versatile Backbone for Dense ... - arXiv
https://arxiv.org/abs/2102.12122
Image 4: arxiv logo>cs> arXiv:2102.12122 Authors:Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao View a PDF of the paper titled Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions, by Wenhai Wang and 8 other authors View a PDF...
Probabilistic Online Event Downsampling
2506.02547v1
cohen2018spatial
\cite{cohen2018spatial}
Spatial and temporal downsampling in event-based visual classification
null
null
true
false
Cohen, Gregory and Afshar, Saeed and Orchard, Garrick and Tapson, Jonathan and Benosman, Ryad and van Schaik, Andre
2,018
null
null
null
IEEE Transactions on Neural Networks and Learning Systems
Spatial and temporal downsampling in event-based visual classification
Spatial and Temporal Downsampling in Event-Based ...
https://www.researchgate.net/publication/322566649_Spatial_and_Temporal_Downsampling_in_Event-Based_Visual_Classification
The results show that both spatial downsampling and temporal downsampling produce improved classification accuracy and, additionally, a lower overall data rate.
Probabilistic Online Event Downsampling
2506.02547v1
ghoshevdownsampling
\cite{ghoshevdownsampling}
EvDownsampling: a robust method for downsampling event camera data
null
null
true
false
Ghosh, Anindya and Nowotny, Thomas and Knight, James
null
null
null
null
null
EvDownsampling: a robust method for downsampling event camera data
a robust method for downsampling event camera data
https://sussex.figshare.com/articles/conference_contribution/EvDownsampling_a_robust_method_for_downsampling_event_camera_data/26970640
by A Ghosh · Cited by 1 — We present a bio-inspired spatio-temporal downsampling technique that can downsample event streams by factors of up to 16 times.
Probabilistic Online Event Downsampling
2506.02547v1
barrios2018less
\cite{barrios2018less}
Less data same information for event-based sensors: A bioinspired filtering and data reduction algorithm
null
null
true
false
Barrios-Avil{\'e}s, Juan and Rosado-Mu{\~n}oz, Alfredo and Medus, Leandro D and Bataller-Mompe{\'a}n, Manuel and Guerrero-Mart{\'\i}nez, Juan F
2,018
null
null
null
Sensors
Less data same information for event-based sensors: A bioinspired filtering and data reduction algorithm
Less Data Same Information for Event-Based Sensors: A Bioinspired ...
https://pmc.ncbi.nlm.nih.gov/articles/PMC6308842/
This work proposes a filtering algorithm (LDSI—Less Data Same Information) which reduces the generated data from event-based sensors without loss of relevant
Probabilistic Online Event Downsampling
2506.02547v1
gupta2020implementing
\cite{gupta2020implementing}
Implementing a foveal-pit inspired filter in a Spiking Convolutional Neural Network: a preliminary study
http://arxiv.org/abs/2105.14326v1
We have presented a Spiking Convolutional Neural Network (SCNN) that incorporates retinal foveal-pit inspired Difference of Gaussian filters and rank-order encoding. The model is trained using a variant of the backpropagation algorithm adapted to work with spiking neurons, as implemented in the Nengo library. We have e...
true
true
Gupta, Shriya TP and Bhattacharya, Basabdatta Sen
2,020
null
null
null
null
Implementing a foveal-pit inspired filter in a Spiking Convolutional Neural Network: a preliminary study
(PDF) Implementing a foveal-pit inspired filter in a Spiking ...
https://www.researchgate.net/publication/352016174_Implementing_a_foveal-pit_inspired_filter_in_a_Spiking_Convolutional_Neural_Network_a_preliminary_study
We have presented a Spiking Convolutional Neural Network (SCNN) that incorporates retinal foveal-pit inspired Difference of Gaussian filters and rank-order
Probabilistic Online Event Downsampling
2506.02547v1
Gruel_2023_WACV
\cite{Gruel_2023_WACV}
Performance Comparison of DVS Data Spatial Downscaling Methods Using Spiking Neural Networks
null
null
true
false
Gruel, Am\'elie and Martinet, Jean and Linares-Barranco, Bernab\'e and Serrano-Gotarredona, Teresa
2,023
January
null
null
null
Performance Comparison of DVS Data Spatial Downscaling Methods Using Spiking Neural Networks
Performance comparison of DVS data spatial downscaling ...
https://openaccess.thecvf.com/content/WACV2023/supplemental/Gruel_Performance_Comparison_of_WACV_2023_supplemental.pdf
Performance comparison of DVS data spatial downscaling methods using Spiking Neural Networks. Supplementary Material. Amélie Gruel. CNRS, i3S, Université Côte
Probabilistic Online Event Downsampling
2506.02547v1
ghosh2023insect
\cite{ghosh2023insect}
Insect-inspired Spatio-temporal Downsampling of Event-based Input
null
null
true
false
Ghosh, Anindya and Nowotny, Thomas and Knight, James C
2,023
null
null
null
null
Insect-inspired Spatio-temporal Downsampling of Event-based Input
Insect-inspired Spatio-temporal Downsampling of Event-based Input
https://dl.acm.org/doi/pdf/10.1145/3589737.3605994
We show that our downsampled event streams achieve high fidelity with a hypothetical low-resolution event camera and improve classification performance on
Probabilistic Online Event Downsampling
2506.02547v1
rizzo2023neuromorphic
\cite{rizzo2023neuromorphic}
Neuromorphic downsampling of event-based camera output
null
null
true
false
Rizzo, Charles P and Schuman, Catherine D and Plank, James S
2,023
null
null
null
null
Neuromorphic downsampling of event-based camera output
Neuromorphic Downsampling of Event-Based Camera Output
https://dl.acm.org/doi/10.1145/3584954.3584962
We construct multiple neuromorphic networks that downsample the camera data so as to make training more effective.
Probabilistic Online Event Downsampling
2506.02547v1
bisulco2020near
\cite{bisulco2020near}
Near-chip Dynamic Vision Filtering for Low-Bandwidth Pedestrian Detection
http://arxiv.org/abs/2004.01689v1
This paper presents a novel end-to-end system for pedestrian detection using Dynamic Vision Sensors (DVSs). We target applications where multiple sensors transmit data to a local processing unit, which executes a detection algorithm. Our system is composed of (i) a near-chip event filter that compresses and denoises th...
true
true
Bisulco, Anthony and Ojeda, Fernando Cladera and Isler, Volkan and Lee, Daniel Dongyuel
2,020
null
null
null
null
Near-chip Dynamic Vision Filtering for Low-Bandwidth Pedestrian Detection
Near-Chip Dynamic Vision Filtering for Low-Bandwidth ...
https://ieeexplore.ieee.org/document/9155035/
by A Bisulco · 2020 · Cited by 12 — This paper presents a novel end-to-end system for pedestrian detection using Dynamic Vision Sensors (DVSs). We target applications where multiple sensors
Probabilistic Online Event Downsampling
2506.02547v1
bi2019graph
\cite{bi2019graph}
Graph-Based Object Classification for Neuromorphic Vision Sensing
http://arxiv.org/abs/1908.06648v1
Neuromorphic vision sensing (NVS)\ devices represent visual information as sequences of asynchronous discrete events (a.k.a., ``spikes'') in response to changes in scene reflectance. Unlike conventional active pixel sensing (APS), NVS allows for significantly higher event sampling rates at substantially increased energ...
true
true
Bi, Yin and Chadha, Aaron and Abbas, Alhabib and Bourtsoulatze, Eirina and Andreopoulos, Yiannis
2,019
null
null
null
null
Graph-Based Object Classification for Neuromorphic Vision Sensing
Graph-based Object Classification for Neuromorphic Vision Sensing
https://github.com/PIX2NVS/NVS2Graph
Our goal is to represent the stream of spike events from neuromorphic vision sensors as a graph and perform convolution on the graph for object classification.
Probabilistic Online Event Downsampling
2506.02547v1
gruel2023frugal
\cite{gruel2023frugal}
Frugal event data: how small is too small? A human performance assessment with shrinking data
null
null
true
false
Gruel, Am{\'e}lie and Carreras, Luc{\'\i}a Trillo and Garc{\'\i}a, Marina Bueno and Kupczyk, Ewa and Martinet, Jean
2,023
null
null
null
null
Frugal event data: how small is too small? A human performance assessment with shrinking data
Frugal event data: how small is too small? A human performance ...
https://ieeexplore.ieee.org/document/10208584/
Frugal event data: how small is too small? A human performance assessment with shrinking data. Abstract: When designing embedded computer vision systems with
Probabilistic Online Event Downsampling
2506.02547v1
araghi2024pushing
\cite{araghi2024pushing}
Pushing the boundaries of event subsampling in event-based video classification using CNNs
http://arxiv.org/abs/2409.08953v1
Event cameras offer low-power visual sensing capabilities ideal for edge-device applications. However, their high event rate, driven by high temporal details, can be restrictive in terms of bandwidth and computational resources. In edge AI applications, determining the minimum amount of events for specific tasks can al...
true
true
Araghi, Hesam and van Gemert, Jan and Tomen, Nergis
2,024
null
null
null
arXiv preprint arXiv:2409.08953
Pushing the boundaries of event subsampling in event-based video classification using CNNs
Pushing the Boundaries of Event Subsampling in Event-Based ...
https://link.springer.com/chapter/10.1007/978-3-031-92460-6_17
In this paper, we study the effect of event subsampling on the accuracy of event data classification using convolutional neural network (CNN) models.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
buda2018systematic
\cite{buda2018systematic}
A systematic study of the class imbalance problem in convolutional neural networks
http://arxiv.org/abs/1710.05381v2
In this study, we systematically investigate the impact of class imbalance on classification performance of convolutional neural networks (CNNs) and compare frequently used methods to address the issue. Class imbalance is a common problem that has been comprehensively studied in classical machine learning, yet very lim...
true
true
Buda, Mateusz and Maki, Atsuto and Mazurowski, Maciej A
2,018
null
null
null
Neural networks
A systematic study of the class imbalance problem in convolutional neural networks
A systematic study of the class imbalance problem in convolutional ...
https://arxiv.org/abs/1710.05381
In this study, we systematically investigate the impact of class imbalance on classification performance of convolutional neural networks (CNNs)
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
byrd2019effect
\cite{byrd2019effect}
What is the Effect of Importance Weighting in Deep Learning?
http://arxiv.org/abs/1812.03372v3
Importance-weighted risk minimization is a key ingredient in many machine learning algorithms for causal inference, domain adaptation, class imbalance, and off-policy reinforcement learning. While the effect of importance weighting is well-characterized for low-capacity misspecified models, little is known about how it...
true
true
Byrd, Jonathon and Lipton, Zachary
2,019
null
null
null
null
What is the Effect of Importance Weighting in Deep Learning?
What is the Effect of Importance Weighting in Deep Learning?
http://arxiv.org/pdf/1812.03372v3
Importance-weighted risk minimization is a key ingredient in many machine learning algorithms for causal inference, domain adaptation, class imbalance, and off-policy reinforcement learning. While the effect of importance weighting is well-characterized for low-capacity misspecified models, little is known about how it...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
drummond2003c4
\cite{drummond2003c4}
C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling
null
null
true
false
Drummond, Chris and Holte, Robert C and others
2,003
null
null
null
null
C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling
[PDF] C4.5, Class Imbalance, and Cost Sensitivity: Why Under-Sampling ...
https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4a57c0ffeec2665caf8e11574ce5a9618304b979
This paper shows that using C4. 5 with under- sampling establishes a reasonable standard for algorithmic comparison.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
pouyanfar2018dynamic
\cite{pouyanfar2018dynamic}
Dynamic sampling in convolutional neural networks for imbalanced data classification
null
null
true
false
Pouyanfar, Samira and Tao, Yudong and Mohan, Anup and Tian, Haiman and Kaseb, Ahmed S and Gauen, Kent and Dailey, Ryan and Aghajanzadeh, Sarah and Lu, Yung-Hsiang and Chen, Shu-Ching and others
2,018
null
null
null
null
Dynamic sampling in convolutional neural networks for imbalanced data classification
Dynamic Sampling in Convolutional Neural Networks ... - IEEE Xplore
https://ieeexplore.ieee.org/document/8396983
Dynamic Sampling in Convolutional Neural Networks for Imbalanced Data Classification | IEEE Conference Publication | IEEE Xplore * Download References) This paper presents a novel model based on the Convolutional Neural Networks (CNNs) to handle such imbalanced and heterogeneous data and successfully identifies the sem...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
shen2016relay
\cite{shen2016relay}
Relay backpropagation for effective learning of deep convolutional neural networks
null
null
true
false
Shen, Li and Lin, Zhouchen and Huang, Qingming
2,016
null
null
null
null
Relay backpropagation for effective learning of deep convolutional neural networks
Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks
http://arxiv.org/pdf/1512.05830v2
Learning deeper convolutional neural networks becomes a tendency in recent years. However, many empirical evidences suggest that performance improvement cannot be gained by simply stacking more layers. In this paper, we consider the issue from an information theoretical perspective, and propose a novel method Relay Bac...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
cui2019class
\cite{cui2019class}
Class-Balanced Loss Based on Effective Number of Samples
http://arxiv.org/abs/1901.05555v1
With the rapid increase of large-scale, real-world datasets, it becomes critical to address the problem of long-tailed data distribution (i.e., a few classes account for most of the data, while most classes are under-represented). Existing solutions typically adopt class re-balancing strategies such as re-sampling and ...
true
true
Cui, Yin and Jia, Menglin and Lin, Tsung-Yi and Song, Yang and Belongie, Serge
2,019
null
null
null
null
Class-Balanced Loss Based on Effective Number of Samples
[PDF] Class-Balanced Loss Based on Effective Number of Samples
https://openaccess.thecvf.com/content_CVPR_2019/papers/Cui_Class-Balanced_Loss_Based_on_Effective_Number_of_Samples_CVPR_2019_paper.pdf
Class-balanced loss uses the effective number of samples, calculated by (1-βn)/(1-β), to re-weight loss, addressing long-tailed data distribution.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
huang2016learning
\cite{huang2016learning}
Learning deep representation for imbalanced classification
null
null
true
false
Huang, Chen and Li, Yining and Loy, Chen Change and Tang, Xiaoou
2,016
null
null
null
null
Learning deep representation for imbalanced classification
[PDF] Learning Deep Representation for Imbalanced Classification
https://openaccess.thecvf.com/content_cvpr_2016/papers/Huang_Learning_Deep_Representation_CVPR_2016_paper.pdf
In this paper, we conduct extensive and systematic experiments to validate the effectiveness of these classic schemes for representa- tion learning on class-
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
wang2017learning
\cite{wang2017learning}
Learning to model the tail
null
null
true
false
Wang, Yu-Xiong and Ramanan, Deva and Hebert, Martial
2,017
null
null
null
Advances in neural information processing systems
Learning to model the tail
Learning to Model the Tail
https://meta-learn.github.io/2017/papers/metalearn17_wang.pdf
by YX Wang · Cited by 850 — We describe an approach to learning from long-tailed, imbalanced datasets that are prevalent in real-world settings. Here, the challenge is to learn
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
khan2017cost
\cite{khan2017cost}
Cost Sensitive Learning of Deep Feature Representations from Imbalanced Data
http://arxiv.org/abs/1508.03422v3
Class imbalance is a common problem in the case of real-world object detection and classification tasks. Data of some classes is abundant making them an over-represented majority, and data of other classes is scarce, making them an under-represented minority. This imbalance makes it challenging for a classifier to appr...
true
true
Khan, Salman H and Hayat, Munawar and Bennamoun, Mohammed and Sohel, Ferdous A and Togneri, Roberto
2,017
null
null
null
IEEE transactions on neural networks and learning systems
Cost Sensitive Learning of Deep Feature Representations from Imbalanced Data
Cost Sensitive Learning of Deep Feature Representations from ...
https://arxiv.org/abs/1508.03422
In this work, we propose a cost sensitive deep neural network which can automatically learn robust feature representations for both the majority and minority
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
shu2019meta
\cite{shu2019meta}
Meta-weight-net: Learning an explicit mapping for sample weighting
null
null
true
false
Shu, Jun and Xie, Qi and Yi, Lixuan and Zhao, Qian and Zhou, Sanping and Xu, Zongben and Meng, Deyu
2,019
null
null
null
Advances in neural information processing systems
Meta-weight-net: Learning an explicit mapping for sample weighting
xjtushujun/meta-weight-net
https://github.com/xjtushujun/meta-weight-net
NeurIPS'19: Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting (Official Pytorch implementation for noisy labels).See more
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
kang2019decoupling
\cite{kang2019decoupling}
Decoupling Representation and Classifier for Long-Tailed Recognition
http://arxiv.org/abs/1910.09217v2
The long-tail distribution of the visual world poses great challenges for deep learning based classification models on how to handle the class imbalance problem. Existing solutions usually involve class-balancing strategies, e.g., by loss re-weighting, data re-sampling, or transfer learning from head- to tail-classes, ...
true
true
Kang, Bingyi and Xie, Saining and Rohrbach, Marcus and Yan, Zhicheng and Gordo, Albert and Feng, Jiashi and Kalantidis, Yannis
2,019
null
null
null
arXiv preprint arXiv:1910.09217
Decoupling Representation and Classifier for Long-Tailed Recognition
[PDF] DECOUPLING REPRESENTATION AND CLASSIFIER - OpenReview
https://openreview.net/pdf?id=r1gRTCVFvB
We evaluate the performance of various sampling and classifier training strategies for long-tailed recognition under both joint and decoupled learning schemes.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
zhou2020bbn
\cite{zhou2020bbn}
BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition
http://arxiv.org/abs/1912.02413v4
Our work focuses on tackling the challenging but natural visual recognition task of long-tailed data distribution (i.e., a few classes occupy most of the data, while most classes have rarely few samples). In the literature, class re-balancing strategies (e.g., re-weighting and re-sampling) are the prominent and effecti...
true
true
Zhou, Boyan and Cui, Quan and Wei, Xiu-Shen and Chen, Zhao-Min
2,020
null
null
null
null
BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition
[PDF] BBN: Bilateral-Branch Network With Cumulative Learning for Long ...
https://openaccess.thecvf.com/content_CVPR_2020/papers/Zhou_BBN_Bilateral-Branch_Network_With_Cumulative_Learning_for_Long-Tailed_Visual_Recognition_CVPR_2020_paper.pdf
Our work focuses on tackling the challenging but natu- ral visual recognition task of long-tailed data distribution. (i.e., a few classes occupy most of the
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
zhang2023deep
\cite{zhang2023deep}
Deep Long-Tailed Learning: A Survey
http://arxiv.org/abs/2110.04596v2
Deep long-tailed learning, one of the most challenging problems in visual recognition, aims to train well-performing deep models from a large number of images that follow a long-tailed class distribution. In the last decade, deep learning has emerged as a powerful recognition model for learning high-quality image repre...
true
true
Zhang, Yifan and Kang, Bingyi and Hooi, Bryan and Yan, Shuicheng and Feng, Jiashi
2,023
null
null
null
IEEE Transactions on Pattern Analysis and Machine Intelligence
Deep Long-Tailed Learning: A Survey
[2110.04596] Deep Long-Tailed Learning: A Survey
https://arxiv.org/abs/2110.04596
by Y Zhang · 2021 · Cited by 819 — This paper aims to provide a comprehensive survey on recent advances in deep long-tailed learning.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
nam2023decoupled
\cite{nam2023decoupled}
Decoupled Training for Long-Tailed Classification With Stochastic Representations
http://arxiv.org/abs/2304.09426v1
Decoupling representation learning and classifier learning has been shown to be effective in classification with long-tailed data. There are two main ingredients in constructing a decoupled learning scheme; 1) how to train the feature extractor for representation learning so that it provides generalizable representatio...
true
true
Nam, Giung and Jang, Sunguk and Lee, Juho
2,023
null
null
null
arXiv preprint arXiv:2304.09426
Decoupled Training for Long-Tailed Classification With Stochastic Representations
Decoupled Training for Long-Tailed Classification With ...
https://arxiv.org/abs/2304.09426
by G Nam · 2023 · Cited by 20 — Abstract:Decoupling representation learning and classifier learning has been shown to be effective in classification with long-tailed data.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
ren2020balanced
\cite{ren2020balanced}
Balanced meta-softmax for long-tailed visual recognition
null
null
true
false
Ren, Jiawei and Yu, Cunjun and Ma, Xiao and Zhao, Haiyu and Yi, Shuai and others
2,020
null
null
null
Advances in neural information processing systems
Balanced meta-softmax for long-tailed visual recognition
Balanced meta-softmax for long-tailed visual recognition
https://dl.acm.org/doi/10.5555/3495724.3496075
In our experiments, we demonstrate that Balanced Meta-Softmax outperforms state-of-the-art long-tailed classification solutions on both visual recognition and
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
menon2020long
\cite{menon2020long}
Long-tail learning via logit adjustment
http://arxiv.org/abs/2007.07314v2
Real-world classification problems typically exhibit an imbalanced or long-tailed label distribution, wherein many labels are associated with only a few samples. This poses a challenge for generalisation on such labels, and also makes na\"ive learning biased towards dominant labels. In this paper, we present two simple...
true
true
Menon, Aditya Krishna and Jayasumana, Sadeep and Rawat, Ankit Singh and Jain, Himanshu and Veit, Andreas and Kumar, Sanjiv
2,020
null
null
null
arXiv preprint arXiv:2007.07314
Long-tail learning via logit adjustment
Long-tail learning via logit adjustment - OpenReview
https://openreview.net/forum?id=37nvvqkCo5
This paper provides a statistical framework for long-tail learning by revisiting the idea of logit adjustment based on the label frequencies.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
cui2021parametric
\cite{cui2021parametric}
Parametric Contrastive Learning
http://arxiv.org/abs/2107.12028v2
In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnabl...
true
true
Cui, Jiequan and Zhong, Zhisheng and Liu, Shu and Yu, Bei and Jia, Jiaya
2,021
null
null
null
null
Parametric Contrastive Learning
Parametric Contrastive Learning
http://arxiv.org/pdf/2107.12028v2
In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnabl...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
cui2023generalized
\cite{cui2023generalized}
Generalized Parametric Contrastive Learning
http://arxiv.org/abs/2209.12400v2
In this paper, we propose the Generalized Parametric Contrastive Learning (GPaCo/PaCo) which works well on both imbalanced and balanced data. Based on theoretical analysis, we observe that supervised contrastive loss tends to bias high-frequency classes and thus increases the difficulty of imbalanced learning. We intro...
true
true
Cui, Jiequan and Zhong, Zhisheng and Tian, Zhuotao and Liu, Shu and Yu, Bei and Jia, Jiaya
2,023
null
null
null
IEEE Transactions on Pattern Analysis and Machine Intelligence
Generalized Parametric Contrastive Learning
Generalized Parametric Contrastive Learning
http://arxiv.org/pdf/2209.12400v2
In this paper, we propose the Generalized Parametric Contrastive Learning (GPaCo/PaCo) which works well on both imbalanced and balanced data. Based on theoretical analysis, we observe that supervised contrastive loss tends to bias high-frequency classes and thus increases the difficulty of imbalanced learning. We intro...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
zhu2022balanced
\cite{zhu2022balanced}
Balanced Contrastive Learning for Long-Tailed Visual Recognition
http://arxiv.org/abs/2207.09052v3
Real-world data typically follow a long-tailed distribution, where a few majority categories occupy most of the data while most minority categories contain a limited number of samples. Classification models minimizing cross-entropy struggle to represent and classify the tail classes. Although the problem of learning un...
true
true
Zhu, Jianggang and Wang, Zheng and Chen, Jingjing and Chen, Yi-Ping Phoebe and Jiang, Yu-Gang
2,022
null
null
null
null
Balanced Contrastive Learning for Long-Tailed Visual Recognition
Balanced Contrastive Learning for Long-Tailed Visual Recognition
https://arxiv.org/abs/2207.09052
The proposed balanced contrastive learning (BCL) method satisfies the condition of forming a regular simplex and assists the optimization of cross-entropy.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
suh2023long
\cite{suh2023long}
Long-Tailed Recognition by Mutual Information Maximization between Latent Features and Ground-Truth Labels
http://arxiv.org/abs/2305.01160v3
Although contrastive learning methods have shown prevailing performance on a variety of representation learning tasks, they encounter difficulty when the training dataset is long-tailed. Many researchers have combined contrastive learning and a logit adjustment technique to address this problem, but the combinations ar...
true
true
Suh, Min-Kook and Seo, Seung-Woo
2,023
null
null
null
arXiv preprint arXiv:2305.01160
Long-Tailed Recognition by Mutual Information Maximization between Latent Features and Ground-Truth Labels
Long-Tailed Recognition by Mutual Information ...
https://openreview.net/pdf?id=KqNX6VOqnJ
by MK Suh · Cited by 27 — Rather, we interpret the long-tailed recognition task as a mutual information maximization between latent features and ground-truth labels. This approach.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
zhu2024generalized
\cite{zhu2024generalized}
Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Models
http://arxiv.org/abs/2310.08106v3
Foundation models like CLIP allow zero-shot transfer on various tasks without additional training data. Yet, the zero-shot performance is less competitive than a fully supervised one. Thus, to enhance the performance, fine-tuning and ensembling are also commonly adopted to better fit the downstream tasks. However, we a...
true
true
Zhu, Beier and Tang, Kaihua and Sun, Qianru and Zhang, Hanwang
2,024
null
null
null
Advances in Neural Information Processing Systems
Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Models
Calibrating Fine-tuned Models by Removing Label Bias in ... - arXiv
https://arxiv.org/abs/2310.08106
Missing: 04/08/2025
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
he2020momentum
\cite{he2020momentum}
Momentum Contrast for Unsupervised Visual Representation Learning
http://arxiv.org/abs/1911.05722v3
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged encoder. This enables building a large and consistent dictionary on-the-fly that facilitates contrastive...
true
true
He, Kaiming and Fan, Haoqi and Wu, Yuxin and Xie, Saining and Girshick, Ross
2,020
null
null
null
null
Momentum Contrast for Unsupervised Visual Representation Learning
Momentum Contrast for Unsupervised Visual Representation Learning
http://arxiv.org/pdf/1911.05722v3
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged encoder. This enables building a large and consistent dictionary on-the-fly that facilitates contrastive...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
chen2020simple
\cite{chen2020simple}
A Simple Framework for Contrastive Learning of Visual Representations
http://arxiv.org/abs/2002.05709v3
This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn use...
true
true
Chen, Ting and Kornblith, Simon and Norouzi, Mohammad and Hinton, Geoffrey
2,020
null
null
null
null
A Simple Framework for Contrastive Learning of Visual Representations
A Simple Framework for Contrastive Learning of Visual Representations
http://arxiv.org/pdf/2002.05709v3
This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn use...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
grill2020bootstrap
\cite{grill2020bootstrap}
Bootstrap your own latent: A new approach to self-supervised Learning
http://arxiv.org/abs/2006.07733v3
We introduce Bootstrap Your Own Latent (BYOL), a new approach to self-supervised image representation learning. BYOL relies on two neural networks, referred to as online and target networks, that interact and learn from each other. From an augmented view of an image, we train the online network to predict the target ne...
true
true
Grill, Jean-Bastien and Strub, Florian and Altch{\'e}, Florent and Tallec, Corentin and Richemond, Pierre and Buchatskaya, Elena and Doersch, Carl and Avila Pires, Bernardo and Guo, Zhaohan and Gheshlaghi Azar, Mohammad and others
2,020
null
null
null
Advances in neural information processing systems
Bootstrap your own latent: A new approach to self-supervised Learning
[PDF] Bootstrap Your Own Latent A New Approach to Self-Supervised ...
https://papers.nips.cc/paper/2020/file/f3ada80d5c4ee70142b17b8192b2958e-Paper.pdf
Richemond∗ ,1,2 Elena Buchatskaya1 , Carl Doersch1 , Bernardo Avila Pires1 , Zhaohan Daniel Guo1 Mohammad Gheshlaghi Azar1, Bilal Piot1, Koray Kavukcuoglu1 , Rémi Munos1 , Michal Valko1 1DeepMind 2Imperial College [jbgrill,fstrub,altche,corentint,richemond]@google.com Abstract We introduce Bootstrap Your Own Latent (BY...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
chen2021exploring
\cite{chen2021exploring}
Exploring Simple Siamese Representation Learning
http://arxiv.org/abs/2011.10566v1
Siamese networks have become a common structure in various recent models for unsupervised visual representation learning. These models maximize the similarity between two augmentations of one image, subject to certain conditions for avoiding collapsing solutions. In this paper, we report surprising empirical results th...
true
true
Chen, Xinlei and He, Kaiming
2,021
null
null
null
null
Exploring Simple Siamese Representation Learning
Exploring Simple Siamese Representation Learning
http://arxiv.org/pdf/2011.10566v1
Siamese networks have become a common structure in various recent models for unsupervised visual representation learning. These models maximize the similarity between two augmentations of one image, subject to certain conditions for avoiding collapsing solutions. In this paper, we report surprising empirical results th...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
khosla2020supervised
\cite{khosla2020supervised}
Supervised Contrastive Learning
http://arxiv.org/abs/2004.11362v5
Contrastive learning applied to self-supervised representation learning has seen a resurgence in recent years, leading to state of the art performance in the unsupervised training of deep image models. Modern batch contrastive approaches subsume or significantly outperform traditional contrastive losses such as triplet...
true
true
Khosla, Prannay and Teterwak, Piotr and Wang, Chen and Sarna, Aaron and Tian, Yonglong and Isola, Phillip and Maschinot, Aaron and Liu, Ce and Krishnan, Dilip
2,020
null
null
null
Advances in neural information processing systems
Supervised Contrastive Learning
[PDF] Supervised Contrastive Learning
https://proceedings.neurips.cc/paper/2020/file/d89a66c7c80a29b1bdbab0f2a1a94af8-Paper.pdf
Supervised contrastive learning uses label information to pull together samples of the same class, unlike self-supervised which uses data augmentations.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
du2024probabilistic
\cite{du2024probabilistic}
Probabilistic Contrastive Learning for Long-Tailed Visual Recognition
http://arxiv.org/abs/2403.06726v2
Long-tailed distributions frequently emerge in real-world data, where a large number of minority categories contain a limited number of samples. Such imbalance issue considerably impairs the performance of standard supervised learning algorithms, which are mainly designed for balanced training sets. Recent investigatio...
true
true
Du, Chaoqun and Wang, Yulin and Song, Shiji and Huang, Gao
2,024
null
null
null
IEEE Transactions on Pattern Analysis and Machine Intelligence
Probabilistic Contrastive Learning for Long-Tailed Visual Recognition
LeapLabTHU/ProCo: [TPAMI 2024] Probabilistic ...
https://github.com/LeapLabTHU/ProCo
GitHub - LeapLabTHU/ProCo: [TPAMI 2024] Probabilistic Contrastive Learning for Long-Tailed Visual Recognition * GitHub Advanced Security Enterprise-grade security features Search code, repositories, users, issues, pull requests... Reload to refresh your session.You signed out in another tab or window. ProCo This repo...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
zhang2022fairness
\cite{zhang2022fairness}
Fairness-aware contrastive learning with partially annotated sensitive attributes
null
null
true
false
Zhang, Fengda and Kuang, Kun and Chen, Long and Liu, Yuxuan and Wu, Chao and Xiao, Jun
2,022
null
null
null
null
Fairness-aware contrastive learning with partially annotated sensitive attributes
Fairness-aware Contrastive Learning with Partially Annotated...
https://openreview.net/forum?id=woa783QMul
The paper proposes a variation of contrastive learning that learns fair representation with partially annotated sensitive attribute labels.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
hou2023subclass
\cite{hou2023subclass}
Subclass-balancing Contrastive Learning for Long-tailed Recognition
http://arxiv.org/abs/2306.15925v2
Long-tailed recognition with imbalanced class distribution naturally emerges in practical machine learning applications. Existing methods such as data reweighing, resampling, and supervised contrastive learning enforce the class balance with a price of introducing imbalance between instances of head class and tail clas...
true
true
Hou, Chengkai and Zhang, Jieyu and Wang, Haonan and Zhou, Tianyi
2,023
null
null
null
null
Subclass-balancing Contrastive Learning for Long-tailed Recognition
Subclass-balancing Contrastive Learning for Long-tailed Recognition
https://arxiv.org/abs/2306.15925
A novel subclass-balancing contrastive learning (SBCL) approach that clusters each head class into multiple subclasses of similar sizes as the tail classes.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
kang2020exploring
\cite{kang2020exploring}
Exploring balanced feature spaces for representation learning
null
null
true
false
Kang, Bingyi and Li, Yu and Xie, Sa and Yuan, Zehuan and Feng, Jiashi
2,020
null
null
null
null
Exploring balanced feature spaces for representation learning
[PDF] EXPLORING BALANCED FEATURE SPACES FOR REP
https://openreview.net/pdf?id=OqtLIabPTit
(4) We develop a new method to explicitly pursue balanced feature spaces for representation learning and it outperforms the popular cross-entropy and
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
li2022targeted
\cite{li2022targeted}
Targeted Supervised Contrastive Learning for Long-Tailed Recognition
http://arxiv.org/abs/2111.13998v2
Real-world data often exhibits long tail distributions with heavy class imbalance, where the majority classes can dominate the training process and alter the decision boundaries of the minority classes. Recently, researchers have investigated the potential of supervised contrastive learning for long-tailed recognition,...
true
true
Li, Tianhong and Cao, Peng and Yuan, Yuan and Fan, Lijie and Yang, Yuzhe and Feris, Rogerio S and Indyk, Piotr and Katabi, Dina
2,022
null
null
null
null
Targeted Supervised Contrastive Learning for Long-Tailed Recognition
[PDF] Targeted Supervised Contrastive Learning for Long-Tailed ...
https://openaccess.thecvf.com/content/CVPR2022/papers/Li_Targeted_Supervised_Contrastive_Learning_for_Long-Tailed_Recognition_CVPR_2022_paper.pdf
TSC is especially effec- tive on long-tailed recognition tasks, since for traditional methods based on supervised contrastive loss, classes with fewer training
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
chen2020big
\cite{chen2020big}
Big Self-Supervised Models are Strong Semi-Supervised Learners
http://arxiv.org/abs/2006.10029v2
One paradigm for learning from few labeled examples while making best use of a large amount of unlabeled data is unsupervised pretraining followed by supervised fine-tuning. Although this paradigm uses unlabeled data in a task-agnostic way, in contrast to common approaches to semi-supervised learning for computer visio...
true
true
Chen, Ting and Kornblith, Simon and Swersky, Kevin and Norouzi, Mohammad and Hinton, Geoffrey E
2,020
null
null
null
Advances in neural information processing systems
Big Self-Supervised Models are Strong Semi-Supervised Learners
[2006.10029] Big Self-Supervised Models are Strong Semi ...
https://arxiv.org/abs/2006.10029
by T Chen · 2020 · Cited by 2883 — We show that it is surprisingly effective for semi-supervised learning on ImageNet. A key ingredient of our approach is the use of big (deep and wide) networks.
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
chen2021empirical
\cite{chen2021empirical}
An Empirical Study of Training Self-Supervised Vision Transformers
http://arxiv.org/abs/2104.02057v4
This paper does not describe a novel method. Instead, it studies a straightforward, incremental, yet must-know baseline given the recent progress in computer vision: self-supervised learning for Vision Transformers (ViT). While the training recipes for standard convolutional networks have been highly mature and robust,...
true
true
Chen, Xinlei and Xie, Saining and He, Kaiming
2,021
null
null
null
null
An Empirical Study of Training Self-Supervised Vision Transformers
[PDF] An Empirical Study of Training Self-Supervised Vision Transformers
https://openaccess.thecvf.com/content/ICCV2021/papers/Chen_An_Empirical_Study_of_Training_Self-Supervised_Vision_Transformers_ICCV_2021_paper.pdf
In summary, we believe that the evidence, challenges, and open questions in this study are worth knowing, if self- supervised Transformers will close the gap in
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
caron2020unsupervised
\cite{caron2020unsupervised}
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
http://arxiv.org/abs/2006.09882v5
Unsupervised image representations have significantly reduced the gap with supervised pretraining, notably with the recent achievements of contrastive learning methods. These contrastive methods typically work online and rely on a large number of explicit pairwise feature comparisons, which is computationally challengi...
true
true
Caron, Mathilde and Misra, Ishan and Mairal, Julien and Goyal, Priya and Bojanowski, Piotr and Joulin, Armand
2,020
null
null
null
Advances in neural information processing systems
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
Unsupervised Learning of Visual Features by Contrasting ...
https://arxiv.org/abs/2006.09882
Authors:Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bojanowski, Armand Joulin View a PDF of the paper titled Unsupervised Learning of Visual Features by Contrasting Cluster Assignments, by Mathilde Caron and 5 other authors Specifically, our method simultaneously clusters the data while enforcing con...
Aligned Contrastive Loss for Long-Tailed Recognition
2506.01071v1
fort2021drawing
\cite{fort2021drawing}
Drawing Multiple Augmentation Samples Per Image During Training Efficiently Decreases Test Error
http://arxiv.org/abs/2105.13343v2
In computer vision, it is standard practice to draw a single sample from the data augmentation procedure for each unique image in the mini-batch. However recent work has suggested drawing multiple samples can achieve higher test accuracies. In this work, we provide a detailed empirical evaluation of how the number of a...
true
true
Fort, Stanislav and Brock, Andrew and Pascanu, Razvan and De, Soham and Smith, Samuel L
2,021
null
null
null
arXiv preprint arXiv:2105.13343
Drawing Multiple Augmentation Samples Per Image During Training Efficiently Decreases Test Error
Drawing Multiple Augmentation Samples Per Image During Training ...
https://www.semanticscholar.org/paper/efcafa65a6e69fa52ceba41d7b6356c17c241edc
It is demonstrated drawing multiple samples per image consistently enhances the test accuracy achieved for both small and large batch training,
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
cao2021video
\cite{cao2021video}
Video Super-Resolution Transformer
http://arxiv.org/abs/2106.06847v3
Video super-resolution (VSR), with the aim to restore a high-resolution video from its corresponding low-resolution version, is a spatial-temporal sequence prediction problem. Recently, Transformer has been gaining popularity due to its parallel computing ability for sequence-to-sequence modeling. Thus, it seems to be ...
true
true
Cao, Jiezhang and Li, Yawei and Zhang, Kai and Van Gool, Luc
2,021
null
null
null
arXiv preprint arXiv:2106.06847
Video Super-Resolution Transformer
[2106.06847] Video Super-Resolution Transformer - arXiv
https://arxiv.org/abs/2106.06847
Abstract:Video super-resolution (VSR), with the aim to restore a high-resolution video from its corresponding low-resolution version,
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
chan2021basicvsr
\cite{chan2021basicvsr}
BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond
http://arxiv.org/abs/2012.02181v2
Video super-resolution (VSR) approaches tend to have more components than the image counterparts as they need to exploit the additional temporal dimension. Complex designs are not uncommon. In this study, we wish to untangle the knots and reconsider some most essential components for VSR guided by four basic functional...
true
true
Chan, Kelvin CK and Wang, Xintao and Yu, Ke and Dong, Chao and Loy, Chen Change
2,021
null
null
null
null
BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond
[PDF] BasicVSR: The Search for Essential Components in Video Super ...
https://openaccess.thecvf.com/content/CVPR2021/papers/Chan_BasicVSR_The_Search_for_Essential_Components_in_Video_Super-Resolution_and_CVPR_2021_paper.pdf
BasicVSR is a video super-resolution pipeline using Propagation, Alignment, Aggregation, and Upsampling, with bidirectional propagation and optical flow for
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
chan2022basicvsr++
\cite{chan2022basicvsr++}
Basicvsr++: Improving video super-resolution with enhanced propagation and alignment
null
null
true
false
Chan, Kelvin CK and Zhou, Shangchen and Xu, Xiangyu and Loy, Chen Change
2,022
null
null
null
null
Basicvsr++: Improving video super-resolution with enhanced propagation and alignment
ckkelvinchan/BasicVSR_PlusPlus: Official repository of " ...
https://github.com/ckkelvinchan/BasicVSR_PlusPlus
GitHub - ckkelvinchan/BasicVSR_PlusPlus: Official repository of "BasicVSR++: Improving Video Super-Resolution with Enhanced Propagation and Alignment" | .dev_scripts/github | .dev_scripts/github | [Feature] Add relevant code (#6) | Apr 18, 2022 | | .github/workflows | .github/workflows | [Feature] Add relevant code (#6...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
isobe2020video
\cite{isobe2020video}
Video Super-Resolution with Recurrent Structure-Detail Network
http://arxiv.org/abs/2008.00455v1
Most video super-resolution methods super-resolve a single reference frame with the help of neighboring frames in a temporal sliding window. They are less efficient compared to the recurrent-based methods. In this work, we propose a novel recurrent video super-resolution method which is both effective and efficient in ...
true
true
Isobe, Takashi and Jia, Xu and Gu, Shuhang and Li, Songjiang and Wang, Shengjin and Tian, Qi
2,020
null
null
null
null
Video Super-Resolution with Recurrent Structure-Detail Network
Video Super-Resolution with Recurrent Structure-Detail Network
https://arxiv.org/abs/2008.00455
We propose a novel recurrent video super-resolution method which is both effective and efficient in exploiting previous frames to super-resolve the current
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
isobe2020video2
\cite{isobe2020video2}
Video Super-resolution with Temporal Group Attention
http://arxiv.org/abs/2007.10595v1
Video super-resolution, which aims at producing a high-resolution video from its corresponding low-resolution version, has recently drawn increasing attention. In this work, we propose a novel method that can effectively incorporate temporal information in a hierarchical way. The input sequence is divided into several ...
true
true
Isobe, Takashi and Li, Songjiang and Jia, Xu and Yuan, Shanxin and Slabaugh, Gregory and Xu, Chunjing and Li, Ya-Li and Wang, Shengjin and Tian, Qi
2,020
null
null
null
null
Video Super-resolution with Temporal Group Attention
[PDF] Video Super-Resolution With Temporal Group Attention
https://openaccess.thecvf.com/content_CVPR_2020/papers/Isobe_Video_Super-Resolution_With_Temporal_Group_Attention_CVPR_2020_paper.pdf
For video super- resolution, both spatial information across positions and temporal information across frames can be used to enhance details for an LR frame.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
isobe2020revisiting
\cite{isobe2020revisiting}
Revisiting temporal modeling for video super-resolution
null
null
true
false
Isobe, Takashi and Zhu, Fang and Jia, Xu and Wang, Shengjin
2,020
null
null
null
arXiv preprint arXiv:2008.05765
Revisiting temporal modeling for video super-resolution
Revisiting Temporal Modeling for Video Super-resolution
https://www.bmvc2020-conference.com/assets/papers/0033.pdf
In this work, we carefully study and compare three temporal modeling methods (2D CNN with early fusion, 3D CNN with slow fusion and Recurrent Neural Network)
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
jo2018deep
\cite{jo2018deep}
Deep video super-resolution network using dynamic upsampling filters without explicit motion compensation
null
null
true
false
Jo, Younghyun and Oh, Seoung Wug and Kang, Jaeyeon and Kim, Seon Joo
2,018
null
null
null
null
Deep video super-resolution network using dynamic upsampling filters without explicit motion compensation
yhjo09/VSR-DUF
https://github.com/yhjo09/VSR-DUF
Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation. This is a tensorflow implementation of the paper.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
liang2024vrt
\cite{liang2024vrt}
VRT: A Video Restoration Transformer
http://arxiv.org/abs/2201.12288v2
Video restoration (e.g., video super-resolution) aims to restore high-quality frames from low-quality frames. Different from single image restoration, video restoration generally requires to utilize temporal information from multiple adjacent but usually misaligned video frames. Existing deep methods generally tackle w...
true
true
Liang, Jingyun and Cao, Jiezhang and Fan, Yuchen and Zhang, Kai and Ranjan, Rakesh and Li, Yawei and Timofte, Radu and Van Gool, Luc
2,024
null
null
null
IEEE Transactions on Image Processing
VRT: A Video Restoration Transformer
VRT: A Video Restoration Transformer
http://arxiv.org/pdf/2201.12288v2
Video restoration (e.g., video super-resolution) aims to restore high-quality frames from low-quality frames. Different from single image restoration, video restoration generally requires to utilize temporal information from multiple adjacent but usually misaligned video frames. Existing deep methods generally tackle w...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
liang2022recurrent
\cite{liang2022recurrent}
Recurrent Video Restoration Transformer with Guided Deformable Attention
http://arxiv.org/abs/2206.02146v3
Video restoration aims at restoring multiple high-quality frames from multiple low-quality frames. Existing video restoration methods generally fall into two extreme cases, i.e., they either restore all frames in parallel or restore the video frame by frame in a recurrent way, which would result in different merits and...
true
true
Liang, Jingyun and Fan, Yuchen and Xiang, Xiaoyu and Ranjan, Rakesh and Ilg, Eddy and Green, Simon and Cao, Jiezhang and Zhang, Kai and Timofte, Radu and Gool, Luc V
2,022
null
null
null
Advances in Neural Information Processing Systems
Recurrent Video Restoration Transformer with Guided Deformable Attention
Recurrent Video Restoration Transformer with Guided Deformable Attention
http://arxiv.org/pdf/2206.02146v3
Video restoration aims at restoring multiple high-quality frames from multiple low-quality frames. Existing video restoration methods generally fall into two extreme cases, i.e., they either restore all frames in parallel or restore the video frame by frame in a recurrent way, which would result in different merits and...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
wang2019edvr
\cite{wang2019edvr}
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
http://arxiv.org/abs/1905.02716v1
Video restoration tasks, including super-resolution, deblurring, etc, are drawing increasing attention in the computer vision community. A challenging benchmark named REDS is released in the NTIRE19 Challenge. This new benchmark challenges existing methods from two aspects: (1) how to align multiple frames given large ...
true
true
Wang, Xintao and Chan, Kelvin CK and Yu, Ke and Dong, Chao and Change Loy, Chen
2,019
null
null
null
null
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
http://arxiv.org/pdf/1905.02716v1
Video restoration tasks, including super-resolution, deblurring, etc, are drawing increasing attention in the computer vision community. A challenging benchmark named REDS is released in the NTIRE19 Challenge. This new benchmark challenges existing methods from two aspects: (1) how to align multiple frames given large ...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
xue2019video
\cite{xue2019video}
Video Enhancement with Task-Oriented Flow
http://arxiv.org/abs/1711.09078v3
Many video enhancement algorithms rely on optical flow to register frames in a video sequence. Precise flow estimation is however intractable; and optical flow itself is often a sub-optimal representation for particular video processing tasks. In this paper, we propose task-oriented flow (TOFlow), a motion representati...
true
true
Xue, Tianfan and Chen, Baian and Wu, Jiajun and Wei, Donglai and Freeman, William T
2,019
null
null
null
International Journal of Computer Vision
Video Enhancement with Task-Oriented Flow
Video Enhancement with Task-Oriented Flow
http://toflow.csail.mit.edu/
In this paper, we propose task-oriented flow (TOFlow), a flow representation tailored for specific video processing tasks.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
liu2013bayesian
\cite{liu2013bayesian}
On Bayesian adaptive video super resolution
null
null
true
false
Liu, Ce and Sun, Deqing
2,013
null
null
null
IEEE transactions on pattern analysis and machine intelligence
On Bayesian adaptive video super resolution
[PDF] On Bayesian Adaptive Video Super Resolution - People
https://people.csail.mit.edu/celiu/pdfs/TPAMI13-VSR.pdf
In this paper, we propose a Bayesian approach to adaptive video super resolution via simultaneously estimating underlying motion, blur kernel and noise level
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
nah2019ntire
\cite{nah2019ntire}
Ntire 2019 challenge on video deblurring and super-resolution: Dataset and study
null
null
true
false
Nah, Seungjun and Baik, Sungyong and Hong, Seokil and Moon, Gyeongsik and Son, Sanghyun and Timofte, Radu and Mu Lee, Kyoung
2,019
null
null
null
null
Ntire 2019 challenge on video deblurring and super-resolution: Dataset and study
[PDF] NTIRE 2019 Challenge on Video Deblurring and Super-Resolution
https://openaccess.thecvf.com/content_CVPRW_2019/papers/NTIRE/Nah_NTIRE_2019_Challenge_on_Video_Deblurring_and_Super-Resolution_Dataset_and_CVPRW_2019_paper.pdf
This paper introduces a novel large dataset for video de- blurring, video super-resolution and studies the state-of- the-art as emerged from the NTIRE 2019
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
yi2019progressive
\cite{yi2019progressive}
Progressive fusion video super-resolution network via exploiting non-local spatio-temporal correlations
null
null
true
false
Yi, Peng and Wang, Zhongyuan and Jiang, Kui and Jiang, Junjun and Ma, Jiayi
2,019
null
null
null
null
Progressive fusion video super-resolution network via exploiting non-local spatio-temporal correlations
Progressive Fusion Video Super-Resolution Network via Exploiting ...
https://github.com/psychopa4/PFNL
GitHub - psychopa4/PFNL: Progressive Fusion Video Super-Resolution Network via Exploiting Non-Local Spatio-Temporal Correlations * GitHub Copilot Write better code with AI * Why GitHub * The ReadME Project GitHub community articles * GitHub Advanced Security Enterprise-grade security features Search code, repos...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
li2020mucan
\cite{li2020mucan}
MuCAN: Multi-Correspondence Aggregation Network for Video Super-Resolution
http://arxiv.org/abs/2007.11803v1
Video super-resolution (VSR) aims to utilize multiple low-resolution frames to generate a high-resolution prediction for each frame. In this process, inter- and intra-frames are the key sources for exploiting temporal and spatial information. However, there are a couple of limitations for existing VSR methods. First, o...
true
true
Li, Wenbo and Tao, Xin and Guo, Taian and Qi, Lu and Lu, Jiangbo and Jia, Jiaya
2,020
null
null
null
null
MuCAN: Multi-Correspondence Aggregation Network for Video Super-Resolution
Multi-Correspondence Aggregation Network for Video Super ... - arXiv
https://arxiv.org/abs/2007.11803
We build an effective multi-correspondence aggregation network (MuCAN) for VSR. Our method achieves state-of-the-art results on multiple benchmark datasets.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
realvsr
\cite{realvsr}
Real-world video super-resolution: A benchmark dataset and a decomposition based learning scheme
null
null
true
false
Yang, Xi and Xiang, Wangmeng and Zeng, Hui and Zhang, Lei
2,021
null
null
null
null
Real-world video super-resolution: A benchmark dataset and a decomposition based learning scheme
IanYeung/RealVSR: Dataset and Code for ICCV 2021 paper "Real ...
https://github.com/IanYeung/RealVSR
Dataset and Code for ICCV 2021 paper "Real-world Video Super-resolution: A Benchmark Dataset and A Decomposition based Learning Scheme"
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
realbasicvsr
\cite{realbasicvsr}
Investigating Tradeoffs in Real-World Video Super-Resolution
http://arxiv.org/abs/2111.12704v1
The diversity and complexity of degradations in real-world video super-resolution (VSR) pose non-trivial challenges in inference and training. First, while long-term propagation leads to improved performance in cases of mild degradations, severe in-the-wild degradations could be exaggerated through propagation, impairi...
true
true
Chan, Kelvin CK and Zhou, Shangchen and Xu, Xiangyu and Loy, Chen Change
2,022
null
null
null
null
Investigating Tradeoffs in Real-World Video Super-Resolution
[PDF] Investigating Tradeoffs in Real-World Video Super-Resolution
https://openaccess.thecvf.com/content/CVPR2022/papers/Chan_Investigating_Tradeoffs_in_Real-World_Video_Super-Resolution_CVPR_2022_paper.pdf
Figure 1. Results on a Real-World Video. In this work, we investigate various tradeoffs caused by the complex and diverse degradations in real-world VSR.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
xie2023mitigating
\cite{xie2023mitigating}
Mitigating Artifacts in Real-World Video Super-Resolution Models
http://arxiv.org/abs/2212.07339v1
The recurrent structure is a prevalent framework for the task of video super-resolution, which models the temporal dependency between frames via hidden states. When applied to real-world scenarios with unknown and complex degradations, hidden states tend to contain unpleasant artifacts and propagate them to restored fr...
true
true
Xie, Liangbin and Wang, Xintao and Shi, Shuwei and Gu, Jinjin and Dong, Chao and Shan, Ying
2,023
null
null
null
null
Mitigating Artifacts in Real-World Video Super-Resolution Models
[PDF] Mitigating Artifacts in Real-World Video Super-resolution Models
https://ojs.aaai.org/index.php/AAAI/article/view/25398/25170
Artifacts in video super-resolution are mitigated by replacing hidden states with a cleaner one using a Hidden State Attention (HSA) module, which uses cheap
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
S4
\cite{S4}
Efficiently modeling long sequences with structured state spaces
null
null
true
false
Gu, Albert and Goel, Karan and R{\'e}, Christopher
2,021
null
null
null
arXiv preprint arXiv:2111.00396
Efficiently modeling long sequences with structured state spaces
Efficiently Modeling Long Sequences with Structured State Spaces
http://arxiv.org/pdf/2111.00396v3
A central goal of sequence modeling is designing a single principled model that can address sequence data across a range of modalities and tasks, particularly on long-range dependencies. Although conventional models including RNNs, CNNs, and Transformers have specialized variants for capturing long dependencies, they s...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
variant1
\cite{variant1}
Long Movie Clip Classification with State-Space Video Models
http://arxiv.org/abs/2204.01692v3
Most modern video recognition models are designed to operate on short video clips (e.g., 5-10s in length). Thus, it is challenging to apply such models to long movie understanding tasks, which typically require sophisticated long-range temporal reasoning. The recently introduced video transformers partially address thi...
true
true
Islam, Md Mohaiminul and Bertasius, Gedas
2,022
null
null
null
null
Long Movie Clip Classification with State-Space Video Models
Long Movie Clip Classification with State-Space Video ...
https://arxiv.org/abs/2204.01692
by MM Islam · 2022 · Cited by 137 — We propose ViS4mer, an efficient long-range video model that combines the strengths of self-attention and the recently introduced structured state-space
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
variant2
\cite{variant2}
S4ND: Modeling Images and Videos as Multidimensional Signals Using State Spaces
http://arxiv.org/abs/2210.06583v2
Visual data such as images and videos are typically modeled as discretizations of inherently continuous, multidimensional signals. Existing continuous-signal models attempt to exploit this fact by modeling the underlying signals of visual (e.g., image) data directly. However, these models have not yet been able to achi...
true
true
Nguyen, Eric and Goel, Karan and Gu, Albert and Downs, Gordon and Shah, Preey and Dao, Tri and Baccus, Stephen and R{\'e}, Christopher
2,022
null
null
null
Advances in neural information processing systems
S4ND: Modeling Images and Videos as Multidimensional Signals Using State Spaces
[PDF] S4ND: Modeling Images and Videos as Multidimensional Signals ...
https://proceedings.neurips.cc/paper_files/paper/2022/file/13388efc819c09564c66ab2dc8463809-Paper-Conference.pdf
S4 investigated state space models, which are linear time-invariant systems that map signals u(t) 7! y(t) and can be represented either as a linear ODE (
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
variant3
\cite{variant3}
Selective structured state-spaces for long-form video understanding
null
null
true
false
Wang, Jue and Zhu, Wentao and Wang, Pichao and Yu, Xiang and Liu, Linda and Omar, Mohamed and Hamid, Raffay
2,023
null
null
null
null
Selective structured state-spaces for long-form video understanding
[PDF] Selective Structured State-Spaces for Long-Form Video ...
https://openaccess.thecvf.com/content/CVPR2023/papers/Wang_Selective_Structured_State-Spaces_for_Long-Form_Video_Understanding_CVPR_2023_paper.pdf
We present extensive comparative results using three challenging long-form video understanding datasets. (LVU, COIN and Breakfast), demonstrating that our ap-.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
variant4
\cite{variant4}
Diagonal State Spaces are as Effective as Structured State Spaces
http://arxiv.org/abs/2203.14343v3
Modeling long range dependencies in sequential data is a fundamental step towards attaining human-level performance in many modalities such as text, vision, audio and video. While attention-based models are a popular and effective choice in modeling short-range interactions, their performance on tasks requiring long ra...
true
true
Gupta, Ankit and Gu, Albert and Berant, Jonathan
2,022
null
null
null
Advances in Neural Information Processing Systems
Diagonal State Spaces are as Effective as Structured State Spaces
Diagonal State Spaces are as Effective as Structured State Spaces
http://arxiv.org/pdf/2203.14343v3
Modeling long range dependencies in sequential data is a fundamental step towards attaining human-level performance in many modalities such as text, vision, audio and video. While attention-based models are a popular and effective choice in modeling short-range interactions, their performance on tasks requiring long ra...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
variant5
\cite{variant5}
Simplified State Space Layers for Sequence Modeling
http://arxiv.org/abs/2208.04933v3
Models using structured state space sequence (S4) layers have achieved state-of-the-art performance on long-range sequence modeling tasks. An S4 layer combines linear state space models (SSMs), the HiPPO framework, and deep learning to achieve high performance. We build on the design of the S4 layer and introduce a new...
true
true
Smith, Jimmy TH and Warrington, Andrew and Linderman, Scott W
2,022
null
null
null
arXiv preprint arXiv:2208.04933
Simplified State Space Layers for Sequence Modeling
Simplified State Space Layers for Sequence Modeling
http://arxiv.org/pdf/2208.04933v3
Models using structured state space sequence (S4) layers have achieved state-of-the-art performance on long-range sequence modeling tasks. An S4 layer combines linear state space models (SSMs), the HiPPO framework, and deep learning to achieve high performance. We build on the design of the S4 layer and introduce a new...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
mamba
\cite{mamba}
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
http://arxiv.org/abs/2312.00752v2
Foundation models, now powering most of the exciting applications in deep learning, are almost universally based on the Transformer architecture and its core attention module. Many subquadratic-time architectures such as linear attention, gated convolution and recurrent models, and structured state space models (SSMs) ...
true
true
Gu, Albert and Dao, Tri
2,023
null
null
null
arXiv preprint arXiv:2312.00752
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
https://openreview.net/forum?id=tEYskw1VY2
This paper proposes Mamba, a linear-time sequence model with an intra-layer combination of Selective S4D, Short Convolution and Gated Linear Unit. The paper
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
domain1
\cite{domain1}
Mamba-nd: Selective state space modeling for multi-dimensional data
null
null
true
false
Li, Shufan and Singh, Harkanwar and Grover, Aditya
2,024
null
null
null
arXiv preprint arXiv:2402.05892
Mamba-nd: Selective state space modeling for multi-dimensional data
Mamba-ND: Selective State Space Modeling for Multi-Dimensional ...
https://arxiv.org/abs/2402.05892
In this work, we present Mamba-ND, a generalized design extending the Mamba architecture to arbitrary multi-dimensional data.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
domain2
\cite{domain2}
PointMamba: A Simple State Space Model for Point Cloud Analysis
http://arxiv.org/abs/2402.10739v5
Transformers have become one of the foundational architectures in point cloud analysis tasks due to their excellent global modeling ability. However, the attention mechanism has quadratic complexity, making the design of a linear complexity method with global modeling appealing. In this paper, we propose PointMamba, tr...
true
true
Liang, Dingkang and Zhou, Xin and Wang, Xinyu and Zhu, Xingkui and Xu, Wei and Zou, Zhikang and Ye, Xiaoqing and Bai, Xiang
2,024
null
null
null
arXiv preprint arXiv:2402.10739
PointMamba: A Simple State Space Model for Point Cloud Analysis
PointMamba: A Simple State Space Model for Point Cloud Analysis
http://arxiv.org/pdf/2402.10739v5
Transformers have become one of the foundational architectures in point cloud analysis tasks due to their excellent global modeling ability. However, the attention mechanism has quadratic complexity, making the design of a linear complexity method with global modeling appealing. In this paper, we propose PointMamba, tr...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
domain3
\cite{domain3}
Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model
http://arxiv.org/abs/2401.09417v3
Recently the state space models (SSMs) with efficient hardware-aware designs, i.e., the Mamba deep learning model, have shown great potential for long sequence modeling. Meanwhile building efficient and generic vision backbones purely upon SSMs is an appealing direction. However, representing visual data is challenging...
true
true
Zhu, Lianghui and Liao, Bencheng and Zhang, Qian and Wang, Xinlong and Liu, Wenyu and Wang, Xinggang
2,024
null
null
null
arXiv preprint arXiv:2401.09417
Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model
Vision Mamba: Efficient Visual Representation Learning with ... - arXiv
https://arxiv.org/abs/2401.09417
Title:Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model View a PDF of the paper titled Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model, by Lianghui Zhu and 5 other authors In this paper, we show that the reliance on self-attention for...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task1
\cite{self-supervised_task1}
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
http://arxiv.org/abs/2006.09882v5
Unsupervised image representations have significantly reduced the gap with supervised pretraining, notably with the recent achievements of contrastive learning methods. These contrastive methods typically work online and rely on a large number of explicit pairwise feature comparisons, which is computationally challengi...
true
true
Caron, Mathilde and Misra, Ishan and Mairal, Julien and Goyal, Priya and Bojanowski, Piotr and Joulin, Armand
2,020
null
null
null
Advances in neural information processing systems
Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
Unsupervised Learning of Visual Features by Contrasting ...
https://arxiv.org/abs/2006.09882
Authors:Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bojanowski, Armand Joulin View a PDF of the paper titled Unsupervised Learning of Visual Features by Contrasting Cluster Assignments, by Mathilde Caron and 5 other authors Specifically, our method simultaneously clusters the data while enforcing con...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task2
\cite{self-supervised_task2}
Emerging Properties in Self-Supervised Vision Transformers
http://arxiv.org/abs/2104.14294v2
In this paper, we question if self-supervised learning provides new properties to Vision Transformer (ViT) that stand out compared to convolutional networks (convnets). Beyond the fact that adapting self-supervised methods to this architecture works particularly well, we make the following observations: first, self-sup...
true
true
Caron, Mathilde and Touvron, Hugo and Misra, Ishan and J{\'e}gou, Herv{\'e} and Mairal, Julien and Bojanowski, Piotr and Joulin, Armand
2,021
null
null
null
null
Emerging Properties in Self-Supervised Vision Transformers
[PDF] Emerging Properties in Self-Supervised Vision Transformers
https://openaccess.thecvf.com/content/ICCV2021/papers/Caron_Emerging_Properties_in_Self-Supervised_Vision_Transformers_ICCV_2021_paper.pdf
Self-supervised ViT features contain semantic segmentation, scene layout, object boundaries, and perform well with k-NN classifiers, unlike supervised ViTs or
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task3
\cite{self-supervised_task3}
A Simple Framework for Contrastive Learning of Visual Representations
http://arxiv.org/abs/2002.05709v3
This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn use...
true
true
Chen, Ting and Kornblith, Simon and Norouzi, Mohammad and Hinton, Geoffrey
2,020
null
null
null
null
A Simple Framework for Contrastive Learning of Visual Representations
A Simple Framework for Contrastive Learning of Visual Representations
http://arxiv.org/pdf/2002.05709v3
This paper presents SimCLR: a simple framework for contrastive learning of visual representations. We simplify recently proposed contrastive self-supervised learning algorithms without requiring specialized architectures or a memory bank. In order to understand what enables the contrastive prediction tasks to learn use...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task4
\cite{self-supervised_task4}
SSR: An Efficient and Robust Framework for Learning with Unknown Label Noise
http://arxiv.org/abs/2111.11288v2
Despite the large progress in supervised learning with neural networks, there are significant challenges in obtaining high-quality, large-scale and accurately labelled datasets. In such a context, how to learn in the presence of noisy labels has received more and more attention. As a relatively complex problem, in orde...
true
true
Feng, Chen and Tzimiropoulos, Georgios and Patras, Ioannis
2,021
null
null
null
arXiv preprint arXiv:2111.11288
SSR: An Efficient and Robust Framework for Learning with Unknown Label Noise
[PDF] SSR: An Efficient and Robust Framework for Learning with Unknown ...
https://bmvc2022.mpi-inf.mpg.de/0372.pdf
In this paper, we consider a novel problem setting, Learning with Unknown. Label Noise (LULN), that is, learning when both the degree and the type of noise are.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task5
\cite{self-supervised_task5}
Self-supervised representation learning with cross-context learning between global and hypercolumn features
null
null
true
false
Gao, Zheng and Feng, Chen and Patras, Ioannis
2,024
null
null
null
null
Self-supervised representation learning with cross-context learning between global and hypercolumn features
[PDF] Self-Supervised Representation Learning With Cross-Context ...
https://openaccess.thecvf.com/content/WACV2024/papers/Gao_Self-Supervised_Representation_Learning_With_Cross-Context_Learning_Between_Global_and_Hypercolumn_WACV_2024_paper.pdf
This leads to a novel self-supervised framework–cross-context learn- ing between global and hypercolumn features (CGH)–that learns representations by
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task6
\cite{self-supervised_task6}
Momentum Contrast for Unsupervised Visual Representation Learning
http://arxiv.org/abs/1911.05722v3
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged encoder. This enables building a large and consistent dictionary on-the-fly that facilitates contrastive...
true
true
He, Kaiming and Fan, Haoqi and Wu, Yuxin and Xie, Saining and Girshick, Ross
2,020
null
null
null
null
Momentum Contrast for Unsupervised Visual Representation Learning
Momentum Contrast for Unsupervised Visual Representation Learning
http://arxiv.org/pdf/1911.05722v3
We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged encoder. This enables building a large and consistent dictionary on-the-fly that facilitates contrastive...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task7
\cite{self-supervised_task7}
Masked Autoencoders Are Scalable Vision Learners
http://arxiv.org/abs/2111.06377v3
This paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image and reconstruct the missing pixels. It is based on two core designs. First, we develop an asymmetric encoder-decoder architecture, with an enco...
true
true
He, Kaiming and Chen, Xinlei and Xie, Saining and Li, Yanghao and Doll{\'a}r, Piotr and Girshick, Ross
2,022
null
null
null
null
Masked Autoencoders Are Scalable Vision Learners
Masked Autoencoders Are Scalable Vision Learners
http://arxiv.org/pdf/2111.06377v3
This paper shows that masked autoencoders (MAE) are scalable self-supervised learners for computer vision. Our MAE approach is simple: we mask random patches of the input image and reconstruct the missing pixels. It is based on two core designs. First, we develop an asymmetric encoder-decoder architecture, with an enco...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
self-supervised_task8
\cite{self-supervised_task8}
SimMIM: A Simple Framework for Masked Image Modeling
http://arxiv.org/abs/2111.09886v2
This paper presents SimMIM, a simple framework for masked image modeling. We simplify recently proposed related approaches without special designs such as block-wise masking and tokenization via discrete VAE or clustering. To study what let the masked image modeling task learn good representations, we systematically st...
true
true
Xie, Zhenda and Zhang, Zheng and Cao, Yue and Lin, Yutong and Bao, Jianmin and Yao, Zhuliang and Dai, Qi and Hu, Han
2,022
null
null
null
null
SimMIM: A Simple Framework for Masked Image Modeling
SimMIM: A Simple Framework for Masked Image Modeling
http://arxiv.org/pdf/2111.09886v2
This paper presents SimMIM, a simple framework for masked image modeling. We simplify recently proposed related approaches without special designs such as block-wise masking and tokenization via discrete VAE or clustering. To study what let the masked image modeling task learn good representations, we systematically st...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
cl1
\cite{cl1}
Masked Siamese Networks for Label-Efficient Learning
http://arxiv.org/abs/2204.07141v1
We propose Masked Siamese Networks (MSN), a self-supervised learning framework for learning image representations. Our approach matches the representation of an image view containing randomly masked patches to the representation of the original unmasked image. This self-supervised pre-training strategy is particularly ...
true
true
Assran, Mahmoud and Caron, Mathilde and Misra, Ishan and Bojanowski, Piotr and Bordes, Florian and Vincent, Pascal and Joulin, Armand and Rabbat, Mike and Ballas, Nicolas
2,022
null
null
null
null
Masked Siamese Networks for Label-Efficient Learning
Masked Siamese Networks for Label-Efficient Learning
https://www.ecva.net/papers/eccv_2022/papers_ECCV/papers/136910442.pdf
by M Assran · Cited by 421 — We propose Masked Siamese Networks (MSNs), a self-supervised learning frame- work that leverages the idea of mask-denoising while avoiding pixel and token-level.
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
cl2
\cite{cl2}
Adaptive Soft Contrastive Learning
http://arxiv.org/abs/2207.11163v1
Self-supervised learning has recently achieved great success in representation learning without human annotations. The dominant method -- that is contrastive learning, is generally based on instance discrimination tasks, i.e., individual samples are treated as independent categories. However, presuming all the samples ...
true
true
Feng, Chen and Patras, Ioannis
2,022
null
null
null
null
Adaptive Soft Contrastive Learning
Adaptive Soft Contrastive Learning
http://arxiv.org/pdf/2207.11163v1
Self-supervised learning has recently achieved great success in representation learning without human annotations. The dominant method -- that is contrastive learning, is generally based on instance discrimination tasks, i.e., individual samples are treated as independent categories. However, presuming all the samples ...
Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution
2506.01037v1
cl3
\cite{cl3}
MaskCon: Masked Contrastive Learning for Coarse-Labelled Dataset
http://arxiv.org/abs/2303.12756v1
Deep learning has achieved great success in recent years with the aid of advanced neural network structures and large-scale human-annotated datasets. However, it is often costly and difficult to accurately and efficiently annotate large-scale datasets, especially for some specialized domains where fine-grained labels a...
true
true
Feng, Chen and Patras, Ioannis
2,023
null
null
null
null
MaskCon: Masked Contrastive Learning for Coarse-Labelled Dataset
Masked Contrastive Learning for Coarse-Labelled Dataset
https://ieeexplore.ieee.org/iel7/10203037/10203050/10203131.pdf
by C Feng · 2023 · Cited by 17 — MaskCon is a contrastive learning method for coarse-labeled datasets, generating soft labels based on sample distances to learn fine-grained representations.