3D Human Pose Estimation On Cmu Panoptic
Metrics
Average MPJPE (mm)
Results
Performance results of various models on this benchmark
Model Name | Average MPJPE (mm) | Paper Title | Repository |
|---|---|---|---|
| Learnable Triangulation of Human Pose | 13.7 | Learnable Triangulation of Human Pose | |
| VTP | 17.62 | VTP: Volumetric Transformer for Multi-view Multi-person 3D Pose Estimation | - |
| MUG | 127.8 | MUG: Multi-human Graph Network for 3D Mesh Reconstruction from 2D Pose | - |
| TesseTrack Multi-View (5 views) | 7.3 | TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking | - |
| TesseTrack Monocular | 18.9 | TesseTrack: End-to-End Learnable Multi-Person Articulated 3D Pose Tracking | - |
| ROMP (ResNet-50) | 127.6 | Monocular, One-stage, Regression of Multiple 3D People | |
| LoCO | 69 | Compressed Volumetric Heatmaps for Multi-Person 3D Pose Estimation | |
| AdaFuse | 13.55 | AdaFuse: Adaptive Multiview Fusion for Accurate Human Pose Estimation in the Wild | |
| InstanceHMR | 126.1 | Instance-aware Contrastive Learning for Occluded Human Mesh Reconstruction |
0 of 9 row(s) selected.