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/KAPAO-L

KAPAO-L

Reported on 33 benchmarks across 3 tasks · 1 paper · 6 SOTA

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

Computer Vision11 results

Pose EstimationonCrowdPose
AP50· 2021-11-16
89.4
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCrowdPose
Test· 2021-11-16
76.6
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
AP· 2021-11-16
70.3
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
AP50· 2021-11-16
91.2
best: 95 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
AP75· 2021-11-16
77.8
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
APL· 2021-11-16
76.8
best: 86.5 (PoseBH-H)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
APM· 2021-11-16
66.3
best: 83.8 (4xRSN-50 (ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCOCO test-dev
AR· 2021-11-16
77.7
best: 88.2 (Simple Pose)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCrowdPose
AP· 2021-11-16
68.9
best: 78.5 (BUCTD-W48 (w/cond. input from PETR, and generative sampling))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCrowdPose
AP75· 2021-11-16
75.6
best: 81.4 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
Pose EstimationonCrowdPose
APM· 2021-11-16
69.9
best: 86.6 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557

Methodology11 results

3DonCrowdPose
AP50· 2021-11-16
89.4
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCrowdPose
Test· 2021-11-16
76.6
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
AP· 2021-11-16
70.3
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
AP50· 2021-11-16
91.2
best: 95 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
AP75· 2021-11-16
77.8
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
APL· 2021-11-16
76.8
best: 86.5 (PoseBH-H)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
APM· 2021-11-16
66.3
best: 83.8 (4xRSN-50 (ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCOCO test-dev
AR· 2021-11-16
77.7
best: 88.2 (Simple Pose)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCrowdPose
AP· 2021-11-16
68.9
best: 78.5 (BUCTD-W48 (w/cond. input from PETR, and generative sampling))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCrowdPose
AP75· 2021-11-16
75.6
best: 81.4 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
3DonCrowdPose
APM· 2021-11-16
69.9
best: 86.6 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557

Audio11 results

1 Image, 2*2 StitchionCrowdPose
AP50· 2021-11-16
89.4
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCrowdPose
Test· 2021-11-16
76.6
SOTA
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
AP· 2021-11-16
70.3
best: 81.1 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
AP50· 2021-11-16
91.2
best: 95 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
AP75· 2021-11-16
77.8
best: 88.2 (ViTPose (ViTAE-G, ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
APL· 2021-11-16
76.8
best: 86.5 (PoseBH-H)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
APM· 2021-11-16
66.3
best: 83.8 (4xRSN-50 (ensemble))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCOCO test-dev
AR· 2021-11-16
77.7
best: 88.2 (Simple Pose)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCrowdPose
AP· 2021-11-16
68.9
best: 78.5 (BUCTD-W48 (w/cond. input from PETR, and generative sampling))
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCrowdPose
AP75· 2021-11-16
75.6
best: 81.4 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557
1 Image, 2*2 StitchionCrowdPose
APM· 2021-11-16
69.9
best: 86.6 (ViTPose-G)
Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation arXiv:2111.08557