desaixie/carve3d

Code for Carve3D: Improving Multi-view Reconstruction Consistency for Diffusion Models with RL Finetuning

18
/ 100
Experimental

This project helps researchers working with 3D content generation from diffusion models to improve the consistency of multi-view reconstructions. It refines existing models that take textual descriptions or multiple 2D images and produce coherent 3D object representations. The primary users are researchers and developers focused on advanced 3D content creation and computer vision.

No commits in the last 6 months.

Use this if you are a researcher or developer aiming to enhance the quality and multi-view consistency of 3D reconstructions generated by diffusion models, particularly for academic benchmarking or advanced research.

Not ideal if you are an end-user looking for a ready-to-use application for casual 3D object generation or someone without a strong background in diffusion models and reinforcement learning.

3D-content-creation computer-vision-research generative-AI 3D-reconstruction machine-learning-research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 3 / 25

How are scores calculated?

Stars

37

Forks

1

Language

Python

License

Last pushed

Jun 16, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/desaixie/carve3d"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.