# Realtime Simulation of Thin-Shell Deformable Materials using CNN-Based Mesh Embedding

arXiv: https://arxiv.org/abs/1909.12354 · HTML: https://mlsanity.com/paper/1909.12354/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Sep 2019.

## Built on by (most cited first, 9 of 9)

- [A Deep Emulator for Secondary Motion of 3D Characters](https://mlsanity.com/paper/2103.01261/) (2021)
- [GLASS: Geometric Latent Augmentation for Shape Spaces](https://mlsanity.com/paper/2108.03225/) (2021)
- [Model reduction for the material point method via an implicit neural representation of the deformation map](https://mlsanity.com/paper/2109.12390/) (2021)
- [Reduced Representation of Deformation Fields for Effective Non-rigid Shape Matching](https://mlsanity.com/paper/2211.14604/) (2022)
- [A Repulsive Force Unit for Garment Collision Handling in Neural Networks](https://mlsanity.com/paper/2207.13871/) (2022)
- [Analytically Integratable Zero-restlength Springs for Capturing Dynamic Modes unrepresented by Quasistatic Neural Networks](https://mlsanity.com/paper/2201.10122/) (2022)
- [A Neural-Network-Based Approach for Loose-Fitting Clothing](https://mlsanity.com/paper/2404.16896/) (2024)
- [Manipulating Elasto-Plastic Objects With 3D Occupancy and Learning-Based Predictive Control](https://mlsanity.com/paper/2505.16249/) (2025)
- [Spatially Adaptive Cloth Regression with Implicit Neural Representations](https://mlsanity.com/paper/2311.16344/) (2023)

## Abstract (arXiv)

We address the problem of accelerating thin-shell deformable object simulations by dimension reduction. We present a new algorithm to embed a high-dimensional configuration space of deformable objects in a low-dimensional feature space, where the configurations of objects and feature points have approximate one-to-one mapping. Our key technique is a graph-based convolutional neural network (CNN) defined on meshes with arbitrary topologies and a new mesh embedding approach based on physics-inspired loss term. We have applied our approach to accelerate high-resolution thin shell simulations corresponding to cloth-like materials, where the configuration space has tens of thousands of degrees of freedom. We show that our physics-inspired embedding approach leads to higher accuracy compared with prior mesh embedding methods. Finally, we show that the temporal evolution of the mesh in the feature space can also be learned using a recurrent neural network (RNN) leading to fully learnable physics simulators. After training our learned simulator runs 500-10000× faster and the accuracy is high enough for robot manipulation tasks.