mlsanityData to 14 Oct 2026

Continuous-Time Cloth Dynamics with Physics-Informed Neural ODEs

arXiv 2610.14342 ↗First posted 9 Oct 2026cs.LG cs.ROcited by 0

Not on the map. Why?

Abstract (arXiv)

We present ContinuousCloth, a continuous-time, physics-informed neural solver for deformable cloth dynamics that enforces a spring--mass ODE within a neural trajectory model. Unlike classical time-stepping simulators that return trajectories only on a discrete integration grid, ContinuousCloth supports querying at arbitrary timestamps and computing displacement, velocity, and acceleration via automatic differentiation. We evaluate ContinuousCloth along two dimensions. In a physics-only setting, ContinuousCloth achieves trajectory accuracy comparable to high-precision solvers across six diverse boundary-condition scenarios while reducing runtime by up to 17.4 times. When fused with sparse and noisy observations, the spring--mass ODE acts as a strong inductive bias for downstream tasks: interpolation and parameter estimation. That is, ContinuousCloth reduces interpolation error by 7.3 times over a data-only neural baseline in simulation and 2.4 times on real-world cloth sequences, and achieves 2.6 times lower parameter error than a differentiable-solver baseline under high-noise conditions, demonstrating consistent robustness under realistic measurement imperfections.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 46 references that are in the map. First posted on arXiv 9 Oct 2026.

Builds on

21 of 46 references.

References · 21

  1. 2020On the eigenvector bias of Fourier feature networks: From regression to solving multi-scale PDEs with physics-informed neural networkscited by 284
  2. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 258
  3. 2023PINNacle: A Comprehensive Benchmark of Physics-Informed Neural Networks for Solving PDEscited by 87
  4. 2025PhysTwin: Physics-Informed Reconstruction and Simulation of Deformable Objects from Videoscited by 87
  5. 2020Physics informed deep learning for computational elastodynamics without labeled datacited by 80
  6. 2020Scalable Differentiable Physics for Learning and Controlcited by 73
  7. 2023Separable Physics-Informed Neural Networkscited by 66
  8. 2021DiffCloth: Differentiable Cloth Simulation with Dry Frictional Contactcited by 65
  9. 2023PINNsFormer: A Transformer-Based Framework For Physics-Informed Neural Networkscited by 64
  10. 2021Learning in Sinusoidal Spaces with Physics-Informed Neural Networkscited by 62
  11. 2022Transfer learning based physics-informed neural networks for solving inverse problems in engineering structures under different loading scenarioscited by 46
  12. 2021Quadratic Residual Networks: A New Class of Neural Networks for Solving Forward and Inverse Problems in Physics Involving PDEscited by 43
9 more
  1. 2024Dynamic 3D Gaussian Tracking for Graph-Based Neural Dynamics Modelingcited by 43
  2. 2022DiffCloud: Real-to-Sim from Point Clouds with Differentiable Simulation and Rendering of Deformable Objectscited by 31
  3. 2025Particle-Grid Neural Dynamics for Learning Deformable Object Models from RGB-D Videoscited by 25
  4. 2025Diffusion Dynamics Models with Generative State Estimation for Cloth Manipulationcited by 19
  5. 2024Unfolding the Literature: A Review of Robotic Cloth Manipulationcited by 17
  6. 2022Physics-Informed Neural Networks for Shell Structurescited by 13
  7. 2024Differentiable Discrete Elastic Rods for Real-Time Modeling of Deformable Linear Objectscited by 10
  8. 2023NeuralClothSim: Neural Deformation Fields Meet the Thin Shell Theorycited by 7
  9. 2025Can Real-to-Sim Approaches Capture Dynamic Fabric Behavior for Robotic Fabric Manipulation?cited by 1

Built on by

Not cited yet.