Learning Deformable Body Interactions With Adaptive Spatial Tokenization

TL;DR

This paper introduces an Adaptive Spatial Tokenization method that enhances the efficiency and scalability of simulating deformable body interactions using attention mechanisms, outperforming existing approaches especially on large-scale meshes.

Contribution

The paper presents a novel tokenization approach that enables scalable and accurate simulation of deformable bodies, overcoming computational challenges of prior graph-based methods.

Findings

Outperforms state-of-the-art methods in deformable body simulation

Effective on large meshes with over 100,000 nodes

Provides a new large-scale dataset for deformable body interactions

Abstract

Simulating interactions between deformable bodies is vital in fields like material science, mechanical design, and robotics. While learning-based methods with Graph Neural Networks (GNNs) are effective at solving complex physical systems, they encounter scalability issues when modeling deformable body interactions. To model interactions between objects, pairwise global edges have to be created dynamically, which is computationally intensive and impractical for large-scale meshes. To overcome these challenges, drawing on insights from geometric representations, we propose an Adaptive Spatial Tokenization (AST) method for efficient representation of physical states. By dividing the simulation space into a grid of cells and mapping unstructured meshes onto this structured grid, our approach naturally groups adjacent mesh nodes. We then apply a cross-attention module to map the sparse cells into a compact, fixed-length embedding, serving as tokens for the entire physical state. Self-attention modules are employed to predict the next state over these tokens in latent space. This framework leverages the efficiency of tokenization and the expressive power of attention mechanisms to achieve accurate and scalable simulation results. Extensive experiments demonstrate that our method significantly outperforms state-of-the-art approaches in modeling deformable body interactions. Notably, it remains effective on large-scale simulations with meshes exceeding 100,000 nodes, where existing methods are hindered by computational limitations. Additionally, we contribute a novel large-scale dataset encompassing a wide range of deformable body interactions to support future research in this area.