Reproduce “Variational r-Adaptive Cloth Simulation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Jiahao Wen, Zhen Chen, Jernej Barbič, Danny M. Kaufman

This research introduces the first r-adaptive method for frictionally contacting cloth dynamics and statics, overcoming fundamental pathologies in thin-shell simulations by preventing element degeneracy and escaping suboptimal physical minima. The approach achieves higher visual fidelity with fewer vertices and less computation time, providing up to 3x speedup for cloth simulations.

Key insights

  • The spatially discretized incremental potential objective in thin-shell ITR is prone to suboptimal local minima and unphysical solutions due to element degeneracy, requiring specific regularization to maintain mesh quality.
  • A non-uniform, degeneracy-activated quality regularization term can effectively stabilize the variational remeshing process by penalizing severely distorted elements without inhibiting beneficial local mesh anisotropies.
  • Computational efficiency in ITR-based methods can be substantially improved by leveraging within-timestep coherence for localized derivative updates and dynamically adjusting solver tolerances.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.