English

PhysSkin: Real-Time and Generalizable Physics-Based Animation via Self-Supervised Neural Skinning

Graphics 2026-05-19 v2 Computer Vision and Pattern Recognition Machine Learning

Abstract

Achieving real-time physics-based animation that generalizes across diverse 3D shapes and discretizations remains a fundamental challenge. We introduce PhysSkin, a physics-informed framework that addresses this challenge. In the spirit of Linear Blend Skinning, we learn continuous skinning fields as basis functions lifting motion subspace coordinates to full-space deformation, with subspace defined by handle transformations. To generate mesh-free, discretization-agnostic, and physically consistent skinning fields that generalize well across diverse 3D shapes, PhysSkin employs a new neural skinning fields autoencoder which consists of a transformer-based encoder and a cross-attention decoder. Furthermore, we also develop a novel physics-informed self-supervised learning strategy that incorporates on-the-fly skinning-field normalization and conflict-aware gradient correction, enabling effective balancing of energy minimization, spatial smoothness, and orthogonality constraints. PhysSkin shows outstanding performance on generalizable neural skinning and enables real-time physics-based animation.

Keywords

Cite

@article{arxiv.2603.23194,
  title  = {PhysSkin: Real-Time and Generalizable Physics-Based Animation via Self-Supervised Neural Skinning},
  author = {Yuanhang Lei and Tao Cheng and Xingxuan Li and Boming Zhao and Siyuan Huang and Ruizhen Hu and Peter Yichen Chen and Hujun Bao and Zhaopeng Cui},
  journal= {arXiv preprint arXiv:2603.23194},
  year   = {2026}
}

Comments

Accepted by CVPR 2026 Highlight. Project Page: https://zju3dv.github.io/PhysSkin/

R2 v1 2026-07-01T11:35:26.822Z