Bayesian Differentiable Physics for Cloth Digitalization

TL;DR

This paper introduces a Bayesian differentiable cloth model that accurately digitalizes cloth materials from limited data, utilizing a new dataset and capturing material heterogeneity effectively.

Contribution

It presents a novel Bayesian differentiable physics model for cloth digitalization that works with small datasets and captures material heterogeneity.

Findings

High accuracy in cloth digitalization

Efficient learning from limited data

Effective modeling of material variations

Abstract

We propose a new method for cloth digitalization. Deviating from existing methods which learn from data captured under relatively casual settings, we propose to learn from data captured in strictly tested measuring protocols, and find plausible physical parameters of the cloths. However, such data is currently absent, so we first propose a new dataset with accurate cloth measurements. Further, the data size is considerably smaller than the ones in current deep learning, due to the nature of the data capture process. To learn from small data, we propose a new Bayesian differentiable cloth model to estimate the complex material heterogeneity of real cloths. It can provide highly accurate digitalization from very limited data samples. Through exhaustive evaluation and comparison, we show our method is accurate in cloth digitalization, efficient in learning from limited data samples, and general in capturing material variations. Code and data are available https://github.com/realcrane/Bayesian-Differentiable-Physics-for-Cloth-Digitalization