Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time

We propose a neural physics system for real-time, interactive fluid simulations. Traditional physics-based methods, while accurate, are computationally intensive and suffer from latency issues. Recent machine‑learning methods reduce computational costs while preserving fidelity; yet most still fail…

关键词：Fluid, Simulation, Animation, Neural Physics,Diffusion Model, Control