PhysiGen: Integrating Collision-Aware Physical Constraints for High-Fidelity Human-Human Interaction Generation

2026-05-01

AI Summary

This work addresses the pervasive issue of interpenetration in generated multi-person interaction motions, which severely compromises realism and usability. The authors propose a lightweight, plug-and-play optimization framework that, for the first time, efficiently integrates collision-aware physical constraints into existing human interaction generation models. By approximating body meshes with geometric primitives for rapid collision detection, the method identifies intersecting regions and explicitly enforces physical plausibility by guiding the optimization toward collision-free configurations. Extensive experiments across multiple datasets and generation models demonstrate that the approach significantly reduces body interpenetration while enhancing both visual coherence and physical validity, outperforming current state-of-the-art methods.

Abstract

Despite substantial progress in text-driven 3D human motion synthesis, generating realistic multi-person interaction sequences remains challenging. Notably, body inter-penetration is a pervasive issue from both data acquisition to the generated results, which significantly undermines the realism and usability. Previous generative models either ignored this issue or introduced computationally expensive mesh-level loss functions to alleviate inter-body collisions. In this paper, we propose a general-purpose and computationally efficient optimization strategy named PhysiGen to explicitly integrate collision-aware physical constraints for human-human interaction generation. Specifically, we simplify the high-resolution human body mesh into geometric primitives to greatly reduce the cost of inter-person collision detection. Moreover, we identify the collision regions as the guidance of the optimization directions. PhysiGen is plug-and-play and can be readily integrated into existing human interaction generation models. Extensive cross-dataset and cross-model experiments show that our method can effectively reduce interpenetration and significantly improve visual coherence and physical plausibility compared to the state-of-the-art methods.

Problem

• human-human interaction
• interpenetration
• physical constraints
• 3D motion generation
• collision avoidance

Innovation

• collision-aware constraints
• human-human interaction generation
• geometric primitives
• interpenetration reduction
• plug-and-play optimization