3D Human Keypoint Estimation on Waymo Open Dataset (WOD) (100% train)
10.1MPJPE (cm)GC-KPL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GC-KPLBackbone=Pre-trained on 200,000 WOD-train, Stage I supervised=true2023.06 | 10.1 | -2.11 | |
| GC-KPLBackbone=Pre-trained on 200,000 WOD-train, Stage I supervised=false2023.06 | 10.6 | -1.61 | |
| GC-KPLBackbone=Pre-trained on 5,000 WOD-train, Stage I supervised=false2023.06 | 10.73 | -1.48 | |
| GC-KPLBackbone=Pre-trained on synthetic only, Stage I supervised=false2023.06 | 11.27 | -0.94 | |
| HUM3DILBackbone=Randomly initialized, Stage I supervised=false2023.06 | 12.21 | — |