Half-Physics: Enabling Kinematic 3D Human Model with Physical Interactions

2025-07-31

AI Summary

Existing kinematic 3D human representations—e.g., based on SMPL-X—lack physically grounded interaction, leading to object interpenetration and implausible dynamics. To address this, we propose a “semi-physical” framework that transforms standard skeletal animations into physically responsive dynamic entities in real time—without learning. Our method tightly couples kinematic driving with rigid-body physics simulation, preserving original pose fidelity and articulatory control while enabling realistic contact, collision, and reaction force responses between the human body and its environment or objects. It generalizes across arbitrary body shapes and motions, requiring no per-subject or per-motion tuning. Evaluated across diverse scenarios, the approach demonstrates physical plausibility, real-time performance (>30 FPS), and zero-penetration robustness. This work establishes a lightweight, general-purpose paradigm for high-fidelity, interactive digital human modeling.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Humans and AI: Interaction Techniques and Devices
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

While current general-purpose 3D human models (e.g., SMPL-X) efficiently represent accurate human shape and pose, they lacks the ability to physically interact with the environment due to the kinematic nature. As a result, kinematic-based interaction models often suffer from issues such as interpenetration and unrealistic object dynamics. To address this limitation, we introduce a novel approach that embeds SMPL-X into a tangible entity capable of dynamic physical interactions with its surroundings. Specifically, we propose a "half-physics" mechanism that transforms 3D kinematic motion into a physics simulation. Our approach maintains kinematic control over inherent SMPL-X poses while ensuring physically plausible interactions with scenes and objects, effectively eliminating penetration and unrealistic object dynamics. Unlike reinforcement learning-based methods, which demand extensive and complex training, our half-physics method is learning-free and generalizes to any body shape and motion; meanwhile, it operates in real time. Moreover, it preserves the fidelity of the original kinematic motion while seamlessly integrating physical interactions

Problem

• Enables 3D human models to physically interact with environments
• Eliminates unrealistic interpenetration and object dynamics issues
• Transforms kinematic motion into real-time physics simulation without training

Innovation

• Embeds SMPL-X into a tangible dynamic entity
• Uses half-physics for kinematic-to-physics transformation
• Learning-free real-time physical interaction solution