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

2026-02-02

AI Summary

Existing methods for extending 3D Gaussian Splatting (3DGS) to 4D physics-aware simulation often suffer from motion distortions due to reliance on manual parameter tuning and neglect of surface structure, resulting in limited generalization and suboptimal optimization efficiency. To address these issues, this work proposes FastPhysGS, a framework that achieves geometry-preserving internal particle initialization through Instance-aware Particle Filling (IPF) combined with Monte Carlo importance sampling. It further introduces Bidirectional Graph Decoupled Optimization (BGDO) to enhance physical simulation efficiency. Additionally, a Vision-Language Model (VLM) is leveraged to adaptively predict material parameters, balancing physical plausibility with geometric fidelity. The proposed method generates high-fidelity dynamic 3DGS scenes in just one minute using only 7GB of GPU memory, significantly outperforming existing approaches.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

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.

Problem

• 3D Gaussian Splatting
• physics-based simulation
• 4D dynamic modeling
• material point method
• surface structure

Innovation

• 3D Gaussian Splatting
• Material Point Method
• Monte Carlo Importance Sampling
• Bidirectional Graph Decoupling Optimization
• Physics-based Simulation