3D Human Pose Estimation on Human3.6M Standard protocol 9 (5 subjects train, 2 subjects test)
54.8Direct MPJPEZhou
Evaluation Results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ZhouBackbone=HourglassNetwork, Camera Model=Weak-perspective, Depth estimation=Additional depth regression sub-network2018.03 | 54.8 | 60.7 | 58.2 | 71.4 | 62 | 53.8 | 55.6 | 75.2 | 111.6 | 64.2 | 65.5 | 66.1 | 63.2 | 51.4 | 55.3 | 64.9 | |
| DepthMapBackbone=HourglassNetwork, Camera Model=Weak-perspective, Depth estimation=N-channel DepthMap encoding2018.03 | 56.6 | 62.6 | 54.7 | 64.4 | 69.7 | 53 | 54.3 | 80.7 | 122.4 | 65.3 | 69.6 | 57.9 | 47 | 65 | 52.5 | 65.77 | |
| MehtaBackbone=Modified ResNet50, Camera Model=Full-perspective, Representation=Map representation2018.03 | 62.6 | 78.1 | 63.4 | 72.5 | 88.3 | 63.1 | 74.8 | 106.6 | 138.7 | 78.8 | 93.8 | 73.9 | 55.8 | 82 | 59.6 | 80.5 |