Measuring Physical Plausibility of 3D Human Poses Using Physics Simulation

2025-02-06

AI Summary

Existing 3D human pose evaluation metrics focus solely on joint position error or isolated physical anomalies (e.g., ground penetration, jitter), failing to assess holistic physical plausibility and dynamic stability throughout motion sequences. This work pioneers the integration of rigid-body physics simulation into pose credibility quantification: we construct a skeletal dynamics model of the human body in PyBullet, jointly simulating gravity, ground contact, and balance constraints. Based on this, we propose two novel metrics—static and dynamic stability scores—that uniformly quantify pose feasibility under real-world physical laws. Experiments demonstrate strong correlation between our metrics and human-perceived motion stability; reveal weak empirical correlation between mainstream heuristic metrics (e.g., MPJPE) and true physical balance; and effectively distinguish the physical plausibility gap between multi-view triangulation baselines and ground-truth poses on Human3.6M.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Modeling humans in physical scenes is vital for understanding human-environment interactions for applications involving augmented reality or assessment of human actions from video (e.g. sports or physical rehabilitation). State-of-the-art literature begins with a 3D human pose, from monocular or multiple views, and uses this representation to ground the person within a 3D world space. While standard metrics for accuracy capture joint position errors, they do not consider physical plausibility of the 3D pose. This limitation has motivated researchers to propose other metrics evaluating jitter, floor penetration, and unbalanced postures. Yet, these approaches measure independent instances of errors and are not representative of balance or stability during motion. In this work, we propose measuring physical plausibility from within physics simulation. We introduce two metrics to capture the physical plausibility and stability of predicted 3D poses from any 3D Human Pose Estimation model. Using physics simulation, we discover correlations with existing plausibility metrics and measuring stability during motion. We evaluate and compare the performances of two state-of-the-art methods, a multi-view triangulated baseline, and ground truth 3D markers from the Human3.6m dataset.

Problem

• Assessing physical plausibility of 3D human poses
• Introducing metrics for stability in motion
• Comparing state-of-the-art 3D pose estimation methods

Innovation

• Physics simulation for plausibility
• Two metrics for stability
• Comparison with Human3.6m dataset