3D Human Pose Estimation On Emdb
평가 지표
Average MPJAE (deg)
Average MPJAE-PA (deg)
Average MPJPE (mm)
Average MPJPE-PA (mm)
Average MVE (mm)
Average MVE-PA (mm)
Jitter (10m/s^3)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Average MPJAE (deg) | Average MPJAE-PA (deg) | Average MPJPE (mm) | Average MPJPE-PA (mm) | Average MVE (mm) | Average MVE-PA (mm) | Jitter (10m/s^3) |
|---|---|---|---|---|---|---|---|
| 3d-human-pose-and-shape-regression-with | 28.4555 | 25.7033 | 131.065 | 82.8502 | 159.956 | 98.1305 | 81.8447 |
| biopose-biomechanically-accurate-3d-pose | - | - | 92.5 | 52.1 | 98.9 | - | - |
| centerhmr-a-bottom-up-single-shot-method-for | 26.5975 | 23.9901 | 112.652 | 75.1869 | 134.863 | 90.648 | 71.2556 |
| cliff-carrying-location-information-in-full | 23.0933 | 21.6265 | 103.134 | 68.7969 | 122.884 | 81.3275 | 55.4525 |
| cross-attention-of-disentangled-modalities | - | - | 108.107 | 66.794 | 119.23 | 81.1847 | 185.933 |
| cross-attention-of-disentangled-modalities | 25.07 | 22.9482 | 115.036 | 72.6765 | 133.566 | 86.0043 | 81.2959 |
| genhmr-generative-human-mesh-recovery | - | - | 88.2 | 51.7 | 99.5 | - | - |
| glamr-global-occlusion-aware-human-mesh | 25.5058 | 23.5484 | 107.797 | 71.0091 | 128.237 | 85.4586 | 67.3522 |
| human-body-model-fitting-by-learned-gradient | 25.1572 | 25.6425 | 115.82 | 81.1389 | 140.63 | 95.7255 | 72.9695 |
| hybrik-a-hybrid-analytical-neural-inverse | 24.5174 | 23.0704 | 103.037 | 65.5935 | 122.193 | 80.3678 | 49.2068 |
| pare-part-attention-regressor-for-3d-human | 24.673 | 22.3842 | 113.887 | 72.203 | 133.247 | 85.3788 | 75.1137 |
| tram-global-trajectory-and-motion-of-3d | - | - | 74.4 | 45.7 | 86.6 | - | - |
| wham-reconstructing-world-grounded-humans | - | - | 79.7 | 50.4 | 94.4 | - | - |