English

3D Shape Generation and Completion through Point-Voxel Diffusion

Computer Vision and Pattern Recognition 2021-08-31 v3

Abstract

We propose a novel approach for probabilistic generative modeling of 3D shapes. Unlike most existing models that learn to deterministically translate a latent vector to a shape, our model, Point-Voxel Diffusion (PVD), is a unified, probabilistic formulation for unconditional shape generation and conditional, multi-modal shape completion. PVD marries denoising diffusion models with the hybrid, point-voxel representation of 3D shapes. It can be viewed as a series of denoising steps, reversing the diffusion process from observed point cloud data to Gaussian noise, and is trained by optimizing a variational lower bound to the (conditional) likelihood function. Experiments demonstrate that PVD is capable of synthesizing high-fidelity shapes, completing partial point clouds, and generating multiple completion results from single-view depth scans of real objects.

Keywords

Cite

@article{arxiv.2104.03670,
  title  = {3D Shape Generation and Completion through Point-Voxel Diffusion},
  author = {Linqi Zhou and Yilun Du and Jiajun Wu},
  journal= {arXiv preprint arXiv:2104.03670},
  year   = {2021}
}

Comments

Project page: https://alexzhou907.github.io/pvd

R2 v1 2026-06-24T00:57:31.206Z