Diffusion Probabilistic Models for 3D Point Cloud Generation

CVPR 2021 Paper: "Diffusion Probabilistic Models for 3D Point Cloud Generation" | Official GitHub

This demo allows you to generate 3D point clouds using pre-trained models.

  • Adjust the Seed for different random initializations.
  • Choose a Model Type (e.g., Airplane, Chair).
  • Control Sampling Flexibility: Lower values tend towards the mean shape, higher values increase diversity.
  • Customize Point Color and Marker Size.

Running on: CPU
Choose Model Type
0 65535
0 2
Point Color
1 10
Examples
Seed Choose Model Type Sampling Flexibility Point Color Marker Size