GPU-Based Simulation of Cloth Wrinkles at Submillimeter Levels

In this paper, we study physics-based cloth simulation in a very high resolution setting, presumably at submillimeter levels with millions of vertices, to meet perceptual precision of our human eyes. State-of-the-art simulation techniques, mostly developed for unstructured triangular meshes, can hardly meet this demand due to their large computational costs and memory footprints. We argue that in a very high resolution, it is more plausible to use regular meshes with an underlying grid structure, which can be highly compatible with GPU acceleration like high-resolution images. Based on this idea, we formulate and solve the nonlinear optimization problem for simulating high-resolution wrinkles, by a fast block-based descent method with reduced memory accesses. We also investigate the development of the collision handling component in our system, whose performance benefits greatly from the grid structure. Finally, we explore various issues related to the applications of our system, including initialization for fast convergence and temporal coherence, gathering effects, inflation and stuffing models, and mesh simplification. We can treat our system as a quasistatic wrinkle synthesis tool, run it as a standalone dynamic simulator, or integrate it into a multi-resolution solver as an additional component. The experiment demonstrates the capability, efficiency and flexibility of our system in producing a variety of high-resolution wrinkles effects.

Automated summary

This paper studies physics-based cloth simulation at a very high resolution, utilizing regular meshes with an underlying grid structure and a fast block-based descent method for simulating high-resolution wrinkles. The collision handling component benefits greatly from the grid structure, and the system shows capability, efficiency, and flexibility in producing a variety of high-resolution wrinkles effects.