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

Autor:	Yan, Han, Zhang, Mingrui, Li, Yang, Ma, Chao, Ji, Pan
Rok vydání:	2024
Předmět:	Computer Science - Computer Vision and Pattern Recognition
Druh dokumentu:	Working Paper
Popis:	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. Comment: Project page: https://wolfball.github.io/phycage/
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2411.18548 Zobrazit plný text záznamu View this record from Arxiv