Command Palette
Search for a command to run...
Multi Person Pose Estimation On Ochuman
평가 지표
AP50
AP75
Validation AP
평가 결과
이 벤치마크에서 각 모델의 성능 결과
| Paper Title | ||||
|---|---|---|---|---|
| MIPNet (gt-bb) | 89.7 | 80.1 | 74.1 | Multi-Instance Pose Networks: Rethinking Top-Down Pose Estimation |
| I²R-Net (1st stage:TransPose-H) | 85 | 72.8 | 67.8 | I^2R-Net: Intra- and Inter-Human Relation Network for Multi-Person Pose Estimation |
| TransPose-H | 82.7 | 67.1 | - | TransPose: Keypoint Localization via Transformer |
| HRFormer-B | 81.4 | 67.1 | 62.1 | HRFormer: High-Resolution Transformer for Dense Prediction |
| SPM | 67.5 | 53.2 | 47.6 | Single-Stage Multi-Person Pose Machines |
| CrowdPose | 40.8 | 29.9 | 27.5 | CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark |
| SimplePose | 37.4 | 26.8 | 24.1 | Simple Baselines for Human Pose Estimation and Tracking |
| Mask R-CNN | 33.2 | 24.5 | 20.2 | Mask R-CNN |
0 of 8 row(s) selected.