Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.08175 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911040637239296 |
|---|---|
| author | Feiz, Hossein Labbé, David Romeas, Thomas Faubert, Jocelyn Andrews, Sheldon |
| author_facet | Feiz, Hossein Labbé, David Romeas, Thomas Faubert, Jocelyn Andrews, Sheldon |
| contents | We propose a novel framework for accurate 3D human pose estimation in combat sports using sparse multi-camera setups. Our method integrates robust multi-view 2D pose tracking via a transformer-based top-down approach, employing epipolar geometry constraints and long-term video object segmentation for consistent identity tracking across views. Initial 3D poses are obtained through weighted triangulation and spline smoothing, followed by kinematic optimization to refine pose accuracy. We further enhance pose realism and robustness by introducing a multi-person physics-based trajectory optimization step, effectively addressing challenges such as rapid motions, occlusions, and close interactions. Experimental results on diverse datasets, including a new benchmark of elite boxing footage, demonstrate state-of-the-art performance. Additionally, we release comprehensive annotated video datasets to advance future research in multi-person pose estimation for combat sports. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_08175 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multi-person Physics-based Pose Estimation for Combat Sports Feiz, Hossein Labbé, David Romeas, Thomas Faubert, Jocelyn Andrews, Sheldon Computer Vision and Pattern Recognition We propose a novel framework for accurate 3D human pose estimation in combat sports using sparse multi-camera setups. Our method integrates robust multi-view 2D pose tracking via a transformer-based top-down approach, employing epipolar geometry constraints and long-term video object segmentation for consistent identity tracking across views. Initial 3D poses are obtained through weighted triangulation and spline smoothing, followed by kinematic optimization to refine pose accuracy. We further enhance pose realism and robustness by introducing a multi-person physics-based trajectory optimization step, effectively addressing challenges such as rapid motions, occlusions, and close interactions. Experimental results on diverse datasets, including a new benchmark of elite boxing footage, demonstrate state-of-the-art performance. Additionally, we release comprehensive annotated video datasets to advance future research in multi-person pose estimation for combat sports. |
| title | Multi-person Physics-based Pose Estimation for Combat Sports |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.08175 |