https://github.com/openai/point-e

Point cloud diffusion for 3D model synthesis
https://github.com/openai/point-e

Last synced: 4 days ago
JSON representation

Point cloud diffusion for 3D model synthesis

• Host: GitHub
• URL: https://github.com/openai/point-e
• Owner: openai
• License: mit
• Created: 2022-12-06T16:32:13.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-07-04T19:05:34.000Z (5 months ago)
• Last Synced: 2024-11-26T05:03:43.586Z (18 days ago)
• Language: Python
• Size: 1.53 MB
• Stars: 6,549
• Watchers: 221
• Forks: 765
• Open Issues: 79
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-ai-tools - Point·E - e development by creating an account on GitHub. (1. <a name='ImageArtGeneration'></a>📷 Image & Art Generation)
• awesome-ChatGPT-repositories - point-e - Point cloud diffusion for 3D model synthesis (Others)
• ai-game-devtools - Point·E
• awesome-generative-ai - openai/point-e
• awesome-chatgpt - Point-E - A System for Generating 3D Point Clouds from Complex Prompts. (Venture Capitalists / OpenAI Products)

README

        
# Point·E

![Animation of four 3D point clouds rotating](point_e/examples/paper_banner.gif)

This is the official code and model release for [Point-E: A System for Generating 3D Point Clouds from Complex Prompts](https://arxiv.org/abs/2212.08751).

# Usage

Install with `pip install -e .`.

To get started with examples, see the following notebooks:

 * [image2pointcloud.ipynb](point_e/examples/image2pointcloud.ipynb) - sample a point cloud, conditioned on some example synthetic view images.

 * [text2pointcloud.ipynb](point_e/examples/text2pointcloud.ipynb) - use our small, worse quality pure text-to-3D model to produce 3D point clouds directly from text descriptions. This model's capabilities are limited, but it does understand some simple categories and colors.

 * [pointcloud2mesh.ipynb](point_e/examples/pointcloud2mesh.ipynb) - try our SDF regression model for producing meshes from point clouds.

For our P-FID and P-IS evaluation scripts, see:

 * [evaluate_pfid.py](point_e/evals/scripts/evaluate_pfid.py)

 * [evaluate_pis.py](point_e/evals/scripts/evaluate_pis.py)

For our Blender rendering code, see [blender_script.py](point_e/evals/scripts/blender_script.py)

# Samples

You can download the seed images and point clouds corresponding to the paper banner images [here](https://openaipublic.azureedge.net/main/point-e/banner_pcs.zip).

You can download the seed images used for COCO CLIP R-Precision evaluations [here](https://openaipublic.azureedge.net/main/point-e/coco_images.zip).