PhyCAGE: Physically Plausible Compositional 3D Asset Generation from a Single Image

TL;DR

PhyCAGE is a novel method that generates physically plausible 3D assets from a single image by combining multi-view image synthesis, Gaussian Splatting, and physics-guided optimization.

Contribution

It introduces PSE-SDS, a new physics-based optimization technique that ensures the physical compatibility of 3D asset components from a single image.

Findings

Successfully generates 3D assets with physical plausibility.

Outperforms existing methods in consistency and realism.

Demonstrates effectiveness on diverse asset types.

Abstract

We present PhyCAGE, the first approach for physically plausible compositional 3D asset generation from a single image. Given an input image, we first generate consistent multi-view images for components of the assets. These images are then fitted with 3D Gaussian Splatting representations. To ensure that the Gaussians representing objects are physically compatible with each other, we introduce a Physical Simulation-Enhanced Score Distillation Sampling (PSE-SDS) technique to further optimize the positions of the Gaussians. It is achieved by setting the gradient of the SDS loss as the initial velocity of the physical simulation, allowing the simulator to act as a physics-guided optimizer that progressively corrects the Gaussians' positions to a physically compatible state. Experimental results demonstrate that the proposed method can generate physically plausible compositional 3D assets given a single image.