Multi-person Physics-based Pose Estimation for Combat Sports
Hossein Feiz, David Labbé, Thomas Romeas, Jocelyn Faubert, Sheldon Andrews · 2025
Read Multi-person Physics-based Pose Estimation for Combat Sports by Hossein Feiz, David Labbé, Thomas Romeas, Jocelyn Faubert, Sheldon Andrews (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-08175.txt.
Description
Hossein Feiz and co-authors build a system that tracks fighters' body poses in three dimensions, in a 2025 paper. Combat sports are a hard test: strikes are fast, views get blocked, and bodies clash at close range. Their pipeline starts with a transformer model that finds each person in each camera view. Geometry rules and video tracking keep each fighter's identity steady across views.
First three-dimensional poses come from weighted triangulation, then get smoothed and refined by a motion model. A final step runs a physics-based cleanup for the whole group, which fights errors from speed and contact. Tests on several data sets, plus a new benchmark of elite boxing film, gave top scores. The team also release labeled video sets for later work.
Catalog record
| CREATOR | Hossein Feiz, David Labbé, Thomas Romeas, Jocelyn Faubert, Sheldon Andrews |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-08175