FastPhysGS: Accelerating Physics-based Dynamic 3DGS Simulation via Interior Completion and Adaptive Optimization

TL;DR

FastPhysGS is a novel framework that accelerates physics-based 3D Gaussian Splatting simulations by combining interior particle filling and adaptive optimization, achieving high fidelity in just one minute.

Contribution

It introduces Interior Completion and Adaptive Optimization techniques that significantly improve efficiency and robustness over prior physics simulation methods.

Findings

Achieves high-fidelity simulation in 1 minute

Uses only 7 GB memory during runtime

Outperforms previous methods in speed and quality

Abstract

Extending 3D Gaussian Splatting (3DGS) to 4D physical simulation remains challenging. Based on the Material Point Method (MPM), existing methods either rely on manual parameter tuning or distill dynamics from video diffusion models, limiting the generalization and optimization efficiency. Recent attempts using LLMs/VLMs suffer from a text/image-to-3D perceptual gap, yielding unstable physics behavior. In addition, they often ignore the surface structure of 3DGS, leading to implausible motion. We propose FastPhysGS, a fast and robust framework for physics-based dynamic 3DGS simulation:(1) Instance-aware Particle Filling (IPF) with Monte Carlo Importance Sampling (MCIS) to efficiently populate interior particles while preserving geometric fidelity; (2) Bidirectional Graph Decoupling Optimization (BGDO), an adaptive strategy that rapidly optimizes material parameters predicted from a VLM. Experiments show FastPhysGS achieves high-fidelity physical simulation in 1 minute using only 7 GB runtime memory, outperforming prior works with broad potential applications.