PG-3DGS: Optimizing 3D Gaussian Splatting to Satisfy Physics Objectives

2026-05-11

AI Summary

This work addresses the limitation of existing 3D Gaussian splatting methods, which prioritize visual fidelity at the expense of physical plausibility, thereby failing to produce functionally viable 3D structures. To bridge this gap, we introduce differentiable physics simulation into the 3D Gaussian splatting optimization framework for the first time, jointly optimizing photometric loss and physics-driven objectives—such as fluid interaction and aerodynamic lift—to achieve both visual realism and physical feasibility. The proposed approach enables end-to-end training and demonstrates significantly enhanced functionality in tasks involving liquid pouring and lift generation. Physical validation via 3D printing confirms that our generated models produce substantially greater aerodynamic lift under identical airflow conditions compared to appearance-only baselines.

Abstract

Recent advances in Gaussian Splatting have enabled fast, high-fidelity 3D scene generation, yet these methods remain purely visual and lack an understanding of how shapes behave in the physical world. We introduce Physics-Guided 3D Gaussian Splatting (PG-3DGS), a framework that couples differentiable physics simulation with 3D Gaussian representations to generate 3D structures satisfying physics functionalities. By allowing physical objectives to guide the shape optimization process alongside visual losses, our approach produces geometries that are not only photometrically accurate but also physically functional. The model learns to adjust shapes so that the generated objects exhibit physically meaningful behaviors, for example, teapots that can pour and airplanes that can generate lift, without sacrificing visual quality. Experiments on pouring and aerodynamic lift tasks show that PG-3DGS improves physical functionality while preserving visual quality. In addition to simulation gains, bench-top physical lift tests with 3D-printed aircraft (Cessna, B-2 Spirit, and paper plane) under identical airflow conditions show higher scale-measured lift for PG-3DGS, generated structures than an appearance-matching baseline in all three cases. Our unified framework connects appearance-based reconstruction with physics-based reasoning, enabling end-to-end generation of 3D structures that both look realistic and function correctly.

Problem

• 3D Gaussian Splatting
• physics functionality
• physical behavior
• shape optimization
• visual fidelity

Innovation

• Physics-Guided Optimization
• 3D Gaussian Splatting
• Differentiable Physics Simulation
• Physically Functional Geometry
• End-to-End 3D Generation