Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/CAP

CAP

Reported on 35 benchmarks across 6 tasks · 4 papers · 20 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision31 results

Image ClassificationonFGVC Aircraft
PARAMS· 2021-01-17
34.2
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Image ClassificationonFood-101
Accuracy· 2021-01-17
98.6
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Image ClassificationonFood-101
PARAMS· 2021-01-17
34.2
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Image ClassificationonCUB-200-2011
Accuracy· 2021-01-17
91.8
best: 92.8 (PIM)
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Fine-Grained Image ClassificationonFGVC Aircraft
PARAMS· 2021-01-17
34.2
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Fine-Grained Image ClassificationonFood-101
Accuracy· 2021-01-17
98.6
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Fine-Grained Image ClassificationonFood-101
PARAMS· 2021-01-17
34.2
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Fine-Grained Image ClassificationonCUB-200-2011
Accuracy· 2021-01-17
91.8
best: 92.8 (PIM)
SOTA
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Person Re-IdentificationonDukeMTMC-reID
Rank-10· 2020-12-19
95.4
best: 97.9 (CTL Model (ResNet50, 256x128))
SOTA
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
MAP· 2020-12-19
67.3
best: 76.8 (TMGF)
SOTA
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
MAP· 2020-12-19
79.2
best: 89.6 (TransReID-SSL (ViTi-S))
SOTA
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
Rank-10· 2020-12-19
97.7
best: 98.7 (TMGF)
SOTA
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Video InpaintingonDAVIS
PSNR· 2019-08-30
30.28
best: 33.82 (DMT)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonDAVIS
SSIM· 2019-08-30
0.9521
best: 0.976 (DMT)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonYouTube-VOS 2018
PSNR· 2019-08-30
31.58
best: 34.43 (ProPainter)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonYouTube-VOS 2018
SSIM· 2019-08-30
0.9607
best: 0.9735 (ProPainter)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Person Re-IdentificationonDukeMTMC-reID
Rank-10· 2021-01-17
91.8
best: 97.9 (CTL Model (ResNet50, 256x128))
Context-aware Attentional Pooling (CAP) for Fine-grained Visual Classification arXiv:2101.06635
Person Re-IdentificationonMarket-1501
Rank-1· 2020-12-19
93.3
best: 98 (st-ReID(RE, RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
Rank-5· 2020-12-19
97.5
best: 98.9 (st-ReID(RE, RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
mAP· 2020-12-19
85.1
best: 96.21 (Unsupervised Pre-training (ResNet101+RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2020-12-19
87.7
best: 95.6 (CTL Model (ResNet50, 256x128))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
Rank-5· 2020-12-19
93.7
best: 96.5 (Viewpoint-Aware Loss(RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
mAP· 2020-12-19
76
best: 97.1 (DenseIL)
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
Rank-1· 2020-12-19
81.1
best: 95.6 (CTL Model (ResNet50, 256x128))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonDukeMTMC-reID
Rank-5· 2020-12-19
89.3
best: 96.5 (Viewpoint-Aware Loss(RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
Rank-1· 2020-12-19
91.4
best: 98 (st-ReID(RE, RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Person Re-IdentificationonMarket-1501
Rank-5· 2020-12-19
96.3
best: 98.9 (st-ReID(RE, RK))
Camera-aware Proxies for Unsupervised Person Re-Identification arXiv:2012.10674
Video InpaintingonDAVIS
Ewarp· 2019-08-30
0.1533
best: 0.1785 (VINet)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonDAVIS
VFID· 2019-08-30
0.182
best: 0.104 (DMT)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonYouTube-VOS 2018
Ewarp· 2019-08-30
0.147
best: 0.1859 (LGTSM)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Video InpaintingonYouTube-VOS 2018
VFID· 2019-08-30
0.071
best: 0.042 (ProPainter)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587

Methodology9 results

3DonDAVIS
PSNR· 2019-08-30
30.28
best: 33.82 (DMT)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonDAVIS
SSIM· 2019-08-30
0.9521
best: 0.976 (DMT)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonYouTube-VOS 2018
PSNR· 2019-08-30
31.58
best: 34.43 (ProPainter)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonYouTube-VOS 2018
SSIM· 2019-08-30
0.9607
best: 0.9735 (ProPainter)
SOTA
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
Anomaly DetectiononOne-class CIFAR-10
AUROC· 2021-12-05
97
best: 99.6 (CLIP (OE))
Constrained Adaptive Projection with Pretrained Features for Anomaly Detection arXiv:2112.02597
3DonDAVIS
Ewarp· 2019-08-30
0.1533
best: 0.1785 (VINet)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonDAVIS
VFID· 2019-08-30
0.182
best: 0.104 (DMT)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonYouTube-VOS 2018
Ewarp· 2019-08-30
0.147
best: 0.1859 (LGTSM)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587
3DonYouTube-VOS 2018
VFID· 2019-08-30
0.071
best: 0.042 (ProPainter)
Copy-and-Paste Networks for Deep Video Inpainting arXiv:1908.11587