[2508.19752] Fast 3D Diffusion for Scalable Granular Media Synthesis

Computer Science > Machine Learning arXiv:2508.19752 (cs) [Submitted on 27 Aug 2025 (v1), last revised 20 Mar 2026 (this version, v2)] Title:Fast 3D Diffusion for Scalable Granular Media Synthesis Authors:Muhammad Moeeze Hassan, Régis Cottereau, Filippo Gatti, Patryk Dec View a PDF of the paper titled Fast 3D Diffusion for Scalable Granular Media Synthesis, by Muhammad Moeeze Hassan and 2 other authors View PDF HTML (experimental) Abstract:Discrete Element Method (DEM) simulations of granular media are computationally intensive, particularly during initialization phases dominated by large displacements and kinetic energy. This paper presents a novel generative pipeline based on 3D diffusion models that directly synthesizes arbitrarily large granular assemblies in mechanically realistic configurations. The approach employs a two-stage pipeline. First, an unconditional diffusion model generates independent 3D voxel grids representing granular media; second, a 3D inpainting model, adapted from 2D techniques using masked inputs and repainting strategies, seamlessly stitches these grids together. The inpainting model uses the outputs of the unconditional diffusion model to learn from the context of adjacent generations and creates new regions that blend smoothly into the context region. Both models are trained on binarized 3D occupancy grids derived from a database of small-scale DEM simulations, scaling linearly with the number of output voxels. Simulations that spanned over d...