Fine-grained differentiable physics: a yarn-level model for fabrics
This work addresses the need for fine-grained models in physics-based learning for composite materials like cloths, offering a novel approach to incorporate sophisticated structures and interactions, though it is incremental in advancing differentiable physics.
The authors tackled the problem of modeling complex physical phenomena like fabrics by developing a fine-grained differentiable physics model at the yarn level, demonstrating high-fidelity dynamics capture and data-efficient learning with meaningful parameter explicability.
Differentiable physics modeling combines physics models with gradient-based learning to provide model explicability and data efficiency. It has been used to learn dynamics, solve inverse problems and facilitate design, and is at its inception of impact. Current successes have concentrated on general physics models such as rigid bodies, deformable sheets, etc., assuming relatively simple structures and forces. Their granularity is intrinsically coarse and therefore incapable of modelling complex physical phenomena. Fine-grained models are still to be developed to incorporate sophisticated material structures and force interactions with gradient-based learning. Following this motivation, we propose a new differentiable fabrics model for composite materials such as cloths, where we dive into the granularity of yarns and model individual yarn physics and yarn-to-yarn interactions. To this end, we propose several differentiable forces, whose counterparts in empirical physics are indifferentiable, to facilitate gradient-based learning. These forces, albeit applied to cloths, are ubiquitous in various physical systems. Through comprehensive evaluation and comparison, we demonstrate our model's explicability in learning meaningful physical parameters, versatility in incorporating complex physical structures and heterogeneous materials, data-efficiency in learning, and high-fidelity in capturing subtle dynamics.