Whole-body pose estimation on COCO-WholeBody (val)
74.4Whole APSapiens-2B
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Sapiens-2BParameters=2B2025.09 | 74.4 | 79.2 | 84.6 | 84.1 | 90.9 | 91.2 | 93.8 | 70.4 | 78.1 | 81 | |
| SapiensParameters=2B2025.09 | 74.4 | — | — | — | — | — | — | 70.4 | 78.1 | 81 | |
| Sapiens-2BModel Variant=Wholebody, Pre-trained Backbone=Sapiens ViT, Params=2.163B2025.09 | 74.4 | — | — | — | — | — | — | — | — | 81 | |
| SDPoseFeature strategy=full feature aggregation2025.09 | 72.8 | 78.6 | 84.1 | 82.3 | 89 | 89.7 | 92.8 | 67.3 | 75.4 | 79.4 | |
| SDPose2025.09 | 72.8 | — | — | — | — | — | — | 67.3 | 75.4 | 79.4 | |
| SDPoseModel Variant=Wholebody, Pre-trained Backbone=Stable Diffusion-v2, Params=0.98B2025.09 | 72.8 | — | — | — | — | — | — | — | — | 79.4 | |
| Sapiens-1BParameters=1B2025.09 | 72.7 | 77.4 | 82.9 | 83 | 89.8 | 90.7 | 93.6 | 69.2 | 77.1 | 79.2 | |
| SapiensParameters=1B2025.09 | 72.7 | — | — | — | — | — | — | 69.2 | 77.1 | 79.2 | |
| Sapiens-1BModel Variant=Wholebody, Pre-trained Backbone=Sapiens ViT, Params=1.169B2025.09 | 72.7 | — | — | — | — | — | — | — | — | 79.2 | |
| SDPose-F2Feature strategy=tuned solely on the second-to-last feature2025.09 | 71.5 | 77.9 | 83.4 | 81.5 | 88.7 | 88.5 | 92.2 | 65.2 | 74 | 78.4 | |
| SDPose-F2Variant=Ablation baseline2025.09 | 71.5 | — | — | — | — | — | — | 65.2 | 74 | 78.4 | |
| Sapiens-0.6BParameters=0.6B2025.09 | 69.5 | 74.3 | 80.2 | 79.4 | 87 | 89.5 | 92.9 | 65.4 | 74 | 76.3 | |
| SapiensParameters=0.6B2025.09 | 69.5 | — | — | — | — | — | — | 65.4 | 74 | 76.3 | |
| DWPose-l2025.09 | 66.5 | 72.2 | 78.9 | 70.4 | 81.7 | 88.7 | 92.1 | 62.1 | 71 | 74.3 | |
| DWPose-lModel Size=l2025.09 | 66.5 | — | — | — | — | — | — | 62.1 | 71 | 74.3 | |
| RTMPose-x2025.09 | 65.3 | 71.4 | 78.4 | 69.2 | 81 | 88.8 | 92.2 | 59 | 68.5 | 73.3 | |
| RTMPose-xModel Size=x2025.09 | 65.3 | — | — | — | — | — | — | 59 | 68.5 | 73.3 | |
| Sapiens-0.3BParameters=0.3B2025.09 | 62 | 66.4 | 73.4 | 67.3 | 78.4 | 87.1 | 91.2 | 58.1 | 67.1 | 69.4 | |
| SapiensParameters=0.3B2025.09 | 62 | — | — | — | — | — | — | 58.1 | 67.1 | 69.4 | |
| VitPose+-HBackbone=ViT-H2025.09 | 61.2 | 75.9 | — | 77.9 | — | 63.6 | — | 54.7 | — | — | |
| VitPose+-HModel Variant=H2025.09 | 61.2 | — | — | — | — | — | — | 54.7 | — | — | |
| VitPose+-LBackbone=ViT-L2025.09 | 60.6 | 75.3 | — | 77.1 | — | 63 | — | 54.2 | — | — | |
| VitPose+-LModel Variant=L2025.09 | 60.6 | — | — | — | — | — | — | 54.2 | — | — | |
| HRNet2025.09 | 58.6 | 70.1 | 77.3 | 58.6 | 69.2 | 72.7 | 78.3 | 51.6 | 60.4 | 67.4 | |
| HRNet2025.09 | 58.6 | — | — | — | — | — | — | 51.6 | 60.4 | 67.4 |