Point-E
Open-source system from OpenAI that generates 3D point clouds and meshes from text prompts or images.
About
Point-E is a research project released by OpenAI that generates 3D point clouds from text descriptions or synthetic images. The system uses a two-stage process: a diffusion model produces a point cloud from the input, then a separate model converts it into a mesh suitable for 3D applications.
It is released under the MIT license and is entirely free to use. Installation is via pip and example notebooks are included. The pure text-to-3D path handles simple objects and basic color descriptions; image-conditioned generation tends to produce higher-quality results. Point-E is aimed at researchers and developers experimenting with text-to-3D pipelines.
It is released under the MIT license and is entirely free to use. Installation is via pip and example notebooks are included. The pure text-to-3D path handles simple objects and basic color descriptions; image-conditioned generation tends to produce higher-quality results. Point-E is aimed at researchers and developers experimenting with text-to-3D pipelines.