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/SPM

SPM

Reported on 53 benchmarks across 6 tasks · 2 papers · 14 SOTA

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

Computer Vision26 results

Source-Free Domain AdaptationonPACS
Average Accuracy· 2025-05-30
86.7
SOTA
Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation arXiv:2505.24216
Pose EstimationonOCHuman
AP50· 2019-08-24
67.5
best: 89.7 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonOCHuman
AP75· 2019-08-24
53.2
best: 80.1 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonOCHuman
Validation AP· 2019-08-24
47.6
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonOCHuman
AP50· 2019-08-24
67.5
best: 89.7 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonOCHuman
AP75· 2019-08-24
53.2
best: 80.1 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonOCHuman
Validation AP· 2019-08-24
47.6
best: 74.1 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Source-Free Domain AdaptationonVisDA-2017
Accuracy· 2025-05-30
89.4
best: 93.2 (RCL)
Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation arXiv:2505.24216
Pose EstimationonCOCO test-dev
AP· 2019-08-24
66.9
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCOCO test-dev
AP50· 2019-08-24
88.5
best: 95 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCOCO test-dev
AP75· 2019-08-24
72.9
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCOCO test-dev
APL· 2019-08-24
73.1
best: 86.5 (PoseBH-H)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCOCO test-dev
APM· 2019-08-24
62.6
best: 83.8 (4xRSN-50 (ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCrowdPose
AP Easy· 2019-08-24
70.3
best: 88.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCrowdPose
AP Hard· 2019-08-24
55.7
best: 466 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCrowdPose
AP Medium· 2019-08-24
64.5
best: 566 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Pose EstimationonCrowdPose
mAP @0.5:0.95· 2019-08-24
63.7
best: 83.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCOCO test-dev
AP· 2019-08-24
66.9
best: 79.2 (SCIO (HRNet-48))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCOCO test-dev
AP50· 2019-08-24
88.5
best: 93.5 (SCIO (HRNet-48))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCOCO test-dev
AP75· 2019-08-24
72.9
best: 85.8 (SCIO (HRNet-48))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCOCO test-dev
APL· 2019-08-24
73.1
best: 84.2 (SCIO (HRNet-48))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCOCO test-dev
APM· 2019-08-24
62.6
best: 74.1 (SCIO (HRNet-48))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCrowdPose
AP Easy· 2019-08-24
70.3
best: 88.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCrowdPose
AP Hard· 2019-08-24
55.7
best: 466 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCrowdPose
AP Medium· 2019-08-24
64.5
best: 566 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Multi-Person Pose EstimationonCrowdPose
mAP @0.5:0.95· 2019-08-24
63.7
best: 83.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220

Methodology15 results

Domain AdaptationonDomainNet
Accuracy· 2025-05-30
71.1
SOTA
Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation arXiv:2505.24216
3DonOCHuman
AP50· 2019-08-24
67.5
best: 89.7 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonOCHuman
AP75· 2019-08-24
53.2
best: 80.1 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonOCHuman
Validation AP· 2019-08-24
47.6
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
Domain AdaptationonPACS
Average Accuracy· 2025-05-30
86.7
best: 99 (SIMPLE+)
Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation arXiv:2505.24216
Domain AdaptationonVisDA-2017
Accuracy· 2025-05-30
89.4
best: 93.2 (RCL)
Shuffle PatchMix Augmentation with Confidence-Margin Weighted Pseudo-Labels for Enhanced Source-Free Domain Adaptation arXiv:2505.24216
3DonCOCO test-dev
AP· 2019-08-24
66.9
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCOCO test-dev
AP50· 2019-08-24
88.5
best: 95 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCOCO test-dev
AP75· 2019-08-24
72.9
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCOCO test-dev
APL· 2019-08-24
73.1
best: 86.5 (PoseBH-H)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCOCO test-dev
APM· 2019-08-24
62.6
best: 83.8 (4xRSN-50 (ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCrowdPose
AP Easy· 2019-08-24
70.3
best: 88.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCrowdPose
AP Hard· 2019-08-24
55.7
best: 466 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCrowdPose
AP Medium· 2019-08-24
64.5
best: 566 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
3DonCrowdPose
mAP @0.5:0.95· 2019-08-24
63.7
best: 83.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220

Audio12 results

1 Image, 2*2 StitchionOCHuman
AP50· 2019-08-24
67.5
best: 89.7 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionOCHuman
AP75· 2019-08-24
53.2
best: 80.1 (MIPNet (gt-bb))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionOCHuman
Validation AP· 2019-08-24
47.6
best: 92.8 (ViTPose (ViTAE-G, GT bounding boxes))
SOTA
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCOCO test-dev
AP· 2019-08-24
66.9
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCOCO test-dev
AP50· 2019-08-24
88.5
best: 95 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCOCO test-dev
AP75· 2019-08-24
72.9
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCOCO test-dev
APL· 2019-08-24
73.1
best: 86.5 (PoseBH-H)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCOCO test-dev
APM· 2019-08-24
62.6
best: 83.8 (4xRSN-50 (ensemble))
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCrowdPose
AP Easy· 2019-08-24
70.3
best: 88.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCrowdPose
AP Hard· 2019-08-24
55.7
best: 466 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCrowdPose
AP Medium· 2019-08-24
64.5
best: 566 (DETRPose-N)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220
1 Image, 2*2 StitchionCrowdPose
mAP @0.5:0.95· 2019-08-24
63.7
best: 83.8 (RTMO-l)
Single-Stage Multi-Person Pose Machines arXiv:1908.09220