Pose Estimation on OCHuman (val)
85.8APBMPv2+ 2x
Evaluation results
| Method | Links | |
|---|---|---|
| BMPv2+ 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 85.8 | |
| BMPv2 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 82.4 | |
| BMPv1 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 79 | |
| PMPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 78.9 | |
| MaskPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 77.1 | |
| ViTPose*-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 75.3 | |
| base SAM 2.1Prompting=Ground-truth pose keypoints2026.01 | 70.2 | |
| Sapiens 0.3bDetection Approach=top-down, Detector=RTMDet-l2026.01 | 69.8 | |
| SAM-pose2segPrompting=Ground-truth pose keypoints, Selection Strategy=MaxSpread32026.01 | 69.5 | |
| ViTPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 66.9 | |
| BMPv2+ 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 55.8 | |
| Pose2SegPrompting=Ground-truth pose keypoints2026.01 | 54.4 | |
| BMPv2 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 51.3 | |
| BUCTD 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 48.8 | |
| BMPv1 2xDetection Approach=top-down, Detector=RTMDet-l, Iterative refinement=2x2026.01 | 48.7 | |
| BUCTDDetection Approach=bottom-up2026.01 | 48.3 | |
| PMPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 47.9 | |
| MaskPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 46.6 | |
| CID-w48Detection Approach=bottom-up2026.01 | 46.1 | |
| ViTPose*-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 44.5 | |
| Sapiens 0.3bDetection Approach=top-down, Detector=RTMDet-l2026.01 | 42 | |
| MIPNetDetection Approach=top-down, Detector=other2026.01 | 42 | |
| DEKRDetection Approach=bottom-up2026.01 | 37.9 | |
| ViTPose-BDetection Approach=top-down, Detector=RTMDet-l2026.01 | 37.6 |