Physics-based Human Pose Estimation from a Single Moving RGB Camera

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Aytekin, Ayce Idil, Li, Chuqiao, Luvizon, Diogo, Dabral, Rishabh, Oswald, Martin, Habermann, Marc, Theobalt, Christian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915406178942976
author Aytekin, Ayce Idil
Li, Chuqiao
Luvizon, Diogo
Dabral, Rishabh
Oswald, Martin
Habermann, Marc
Theobalt, Christian
author_facet Aytekin, Ayce Idil
Li, Chuqiao
Luvizon, Diogo
Dabral, Rishabh
Oswald, Martin
Habermann, Marc
Theobalt, Christian
contents Most monocular and physics-based human pose tracking methods, while achieving state-of-the-art results, suffer from artifacts when the scene does not have a strictly flat ground plane or when the camera is moving. Moreover, these methods are often evaluated on in-the-wild real world videos without ground-truth data or on synthetic datasets, which fail to model the real world light transport, camera motion, and pose-induced appearance and geometry changes. To tackle these two problems, we introduce MoviCam, the first non-synthetic dataset containing ground-truth camera trajectories of a dynamically moving monocular RGB camera, scene geometry, and 3D human motion with human-scene contact labels. Additionally, we propose PhysDynPose, a physics-based method that incorporates scene geometry and physical constraints for more accurate human motion tracking in case of camera motion and non-flat scenes. More precisely, we use a state-of-the-art kinematics estimator to obtain the human pose and a robust SLAM method to capture the dynamic camera trajectory, enabling the recovery of the human pose in the world frame. We then refine the kinematic pose estimate using our scene-aware physics optimizer. From our new benchmark, we found that even state-of-the-art methods struggle with this inherently challenging setting, i.e. a moving camera and non-planar environments, while our method robustly estimates both human and camera poses in world coordinates.
format Preprint
id arxiv_https___arxiv_org_abs_2507_17406
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Physics-based Human Pose Estimation from a Single Moving RGB Camera
Aytekin, Ayce Idil
Li, Chuqiao
Luvizon, Diogo
Dabral, Rishabh
Oswald, Martin
Habermann, Marc
Theobalt, Christian
Computer Vision and Pattern Recognition
Most monocular and physics-based human pose tracking methods, while achieving state-of-the-art results, suffer from artifacts when the scene does not have a strictly flat ground plane or when the camera is moving. Moreover, these methods are often evaluated on in-the-wild real world videos without ground-truth data or on synthetic datasets, which fail to model the real world light transport, camera motion, and pose-induced appearance and geometry changes. To tackle these two problems, we introduce MoviCam, the first non-synthetic dataset containing ground-truth camera trajectories of a dynamically moving monocular RGB camera, scene geometry, and 3D human motion with human-scene contact labels. Additionally, we propose PhysDynPose, a physics-based method that incorporates scene geometry and physical constraints for more accurate human motion tracking in case of camera motion and non-flat scenes. More precisely, we use a state-of-the-art kinematics estimator to obtain the human pose and a robust SLAM method to capture the dynamic camera trajectory, enabling the recovery of the human pose in the world frame. We then refine the kinematic pose estimate using our scene-aware physics optimizer. From our new benchmark, we found that even state-of-the-art methods struggle with this inherently challenging setting, i.e. a moving camera and non-planar environments, while our method robustly estimates both human and camera poses in world coordinates.
title Physics-based Human Pose Estimation from a Single Moving RGB Camera
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.17406