TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos
About
We propose TRAM, a two-stage method to reconstruct a human's global trajectory and motion from in-the-wild videos. TRAM robustifies SLAM to recover the camera motion in the presence of dynamic humans and uses the scene background to derive the motion scale. Using the recovered camera as a metric-scale reference frame, we introduce a video transformer model (VIMO) to regress the kinematic body motion of a human. By composing the two motions, we achieve accurate recovery of 3D humans in the world space, reducing global motion errors by a large margin from prior work. https://yufu-wang.github.io/tram4d/
Yufu Wang, Ziyun Wang, Lingjie Liu, Kostas Daniilidis• 2024
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| 3D Human Pose and Shape Estimation | EMDB Protocol 1 24 joints | PA-MPJPE45.7 | 31 | |
| Human Mesh Reconstruction | 3DPW 14 joints (test) | PA-MPJPE35.6 | 26 | |
| Human Mesh Reconstruction | EMDB 24 joints (test) | PA-MPJPE45.7 | 21 | |
| Human Mesh Recovery | 3DPW 14 (test) | PA-MPJPE35.6 | 10 | |
| Global human motion estimation | EMDB 2 | WA-MPJPE76.4 | 8 | |
| Global human motion estimation | RICH | WA-MPJPE127.8 | 7 | |
| Human global trajectory and motion reconstruction | EMDB 2 | PA-MPJPE38.1 | 5 | |
| Human motion estimation in world coordinates | EMDB-2 24 joints (test) | WA-MPJPE76.4 | 4 | |
| Human Trajectory Reconstruction | SLOPER4D (test) | WA-MPJPE149.5 | 4 | |
| Scene Geometry Reconstruction | SLOPER4D (test) | Chamfer Distance10.66 | 3 |
Showing 10 of 10 rows