RapidPoseTriangulation: Multi-view Multi-person Whole-body Human Pose Triangulation in a Millisecond
About
The integration of multi-view imaging and pose estimation represents a significant advance in computer vision applications, offering new possibilities for understanding human movement and interactions. This work presents a new algorithm that improves multi-view multi-person pose estimation, focusing on fast triangulation speeds and good generalization capabilities. The approach extends to whole-body pose estimation, capturing details from facial expressions to finger movements across multiple individuals and viewpoints. Adaptability to different settings is demonstrated through strong performance across unseen datasets and configurations. To support further progress in this field, all of this work is publicly accessible.
Daniel Bermuth, Alexander Poeppel, Wolfgang Reif• 2025
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| 3D Human Pose Estimation | Campus | PCP95.2 | 36 | |
| 3D Human Pose Estimation | Shelf (test) | -- | 27 | |
| 3D Human Pose Estimation | Chi3D | Invalid Rate0.00e+0 | 14 | |
| 3D Human Pose Estimation | shelf | Latency (ms)0.1 | 11 | |
| 3D Multi-person Pose Estimation | human36m, shelf, campus, mvor, chi3d, tsinghua Averaged generalization (test) | PCP91.3 | 10 | |
| 3D Multi-person Pose Estimation | Tsinghua (first sequence) | PCP98.7 | 10 | |
| 3D Human Pose Estimation | Panoptic transfer without depth | PCP99.1 | 5 | |
| 3D Multi-person Pose Estimation | egohumans legoassemble 1.0 (test) | PCP100 | 4 | |
| 3D Human Pose Estimation | MVOR (transfer) | PCP59 | 4 | |
| 3D Multi-person Pose Estimation | egohumans tennis 1.0 (test) | PCP99.9 | 3 |
Showing 10 of 17 rows