nv-tlabs/cosmos-transfer1-diffusion-renderer

Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

43
/ 100
Emerging

This project helps operations engineers and AI developers adjust lighting in existing images and videos without re-shooting them. You input existing footage, and it outputs new versions of that footage with different lighting, or even extracts detailed material and depth maps. This is ideal for training AI models to perform reliably in varied real-world lighting conditions.

786 stars. No commits in the last 6 months.

Use this if you need to create diverse training datasets for AI vision systems by altering illumination in existing video or image assets.

Not ideal if you are looking for simple, artistic video color grading or basic photo editing for social media.

AI training data synthetic data generation computer vision robotics simulation perception model development
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 16 / 25

How are scores calculated?

Stars

786

Forks

59

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Oct 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/nv-tlabs/cosmos-transfer1-diffusion-renderer"

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