A Physics-embedded Deep Learning Framework for Cloth Simulation

arXiv:2403.12820v1 Announce Type: cross
Abstract: Delicate cloth simulations have long been desired in computer graphics. Various methods were proposed to improve engaged force interactions, collision handling, and numerical integrations. Deep learning has the potential to achieve fast and real-time simulation, but common neural network structures often demand many parameters to capture cloth dynamics. This paper proposes a physics-embedded learning framework that directly encodes physical features of cloth simulation. The convolutional neural network is used to represent spatial correlations of the …

abstract arxiv collision computer computer graphics cs.gr cs.lg deep learning deep learning framework demand embedded framework graphics integrations interactions network neural network numerical parameters physics real-time simulation simulations type