PhysMAS improves realistic 4D dynamic scene generation with multi-agent physics approach
PhysMAS: Physics-Grounded Multi-Agent Synthesis of Compositional 4D Gaussians
Artificial Intelligence
Summary
Creating realistic 3D animations of moving objects that behave according to physics is challenging, especially when multiple parts or objects interact. The authors present PhysMAS, a system that uses several specialized agents to assign physical properties and simulate motion based on input views and motion instructions. This method avoids slow trial-and-error adjustments and better keeps the different parts correctly identified and physically plausible. Their tests show that PhysMAS produces scenes that look more realistic and run faster than previous physics-based methods.
4D Gaussian synthesisMaterial Point Method (MPM)physical simulationdynamic scene generationmulti-agent systemsscore distillation samplingmotion promptsemantic alignmentforward simulationdiffusion models
Authors
Jiang Qin, Chunji Lv, Yangguang Wei, Yang Gao, Ming Liu, Lizhong Ding, Ye Yuan, Yinjie Lei, Changsheng Li
Abstract
Efficient, fully automatic, and physically plausible 4D Gaussian synthesis is an important goal for dynamic scene generation. Recent physics-based methods couple 3D Gaussians with the Material Point Method (MPM) to generate physically driven motion, but extending this paradigm to heterogeneous multi-part objects and interacting multi-object scenes remains challenging. Object-level physical assignment collapses distinct parts into a single material state, while one-shot predictions from large language models, vision-language models, or agents neither reliably bind different materials to identified parts nor verify that the resulting MPM configuration is executable. Score Distillation Sampling (SDS)-based parameter optimization, meanwhile, requires repeated per-scene score evaluations and gradient backpropagation, incurring lengthy optimization and potentially yielding suboptimal or unstable solutions. We therefore present PhysMAS, a physics-grounded multi-agent framework. From a motion prompt and four scene views, an Object-Part Scene Agent establishes persistent identities and calls a Material Reasoning Agent for part-wise profiles. It invokes solver-aware skills to bind these identities and profiles to per-particle MPM fields and execute all objects in a shared domain; the framework then screens candidate forward-simulation results. This supports heterogeneous multi-part and interacting multi-object scenes without per-scene diffusion-score backpropagation. Extensive experiments demonstrate that, compared with recent physics-based 4D Gaussian baselines that rely on SDS, PhysMAS achieves better semantic alignment and perceived physical plausibility while requiring less runtime.