duxiaodan/intrinsic-lora

Official repo of ๐—š๐—ฒ๐—ป๐—ฒ๐—ฟ๐—ฎ๐˜๐—ถ๐˜ƒ๐—ฒ ๐— ๐—ผ๐—ฑ๐—ฒ๐—น๐˜€: ๐—ช๐—ต๐—ฎ๐˜ ๐——๐—ผ ๐—ง๐—ต๐—ฒ๐˜† ๐—ž๐—ป๐—ผ๐˜„? ๐——๐—ผ ๐—ง๐—ต๐—ฒ๐˜† ๐—ž๐—ป๐—ผ๐˜„ ๐—ง๐—ต๐—ถ๐—ป๐—ด๐˜€? ๐—Ÿ๐—ฒ๐˜'๐˜€ ๐—™๐—ถ๐—ป๐—ฑ ๐—ข๐˜‚๐˜! (Previously titled ๐˜๐˜ฏ๐˜ต๐˜ณ๐˜ช๐˜ฏ๐˜ด๐˜ช๐˜ค ๐˜“๐˜ฐ๐˜™๐˜ˆ: ๐˜ˆ ๐˜Ž๐˜ฆ๐˜ฏ๐˜ฆ๐˜ณ๐˜ข๐˜ญ๐˜ช๐˜ด๐˜ต ๐˜ˆ๐˜ฑ๐˜ฑ๐˜ณ๐˜ฐ๐˜ข๐˜ค๐˜ฉ ๐˜ง๐˜ฐ๐˜ณ ๐˜‹๐˜ช๐˜ด๐˜ค๐˜ฐ๐˜ท๐˜ฆ๐˜ณ๐˜ช๐˜ฏ๐˜จ ๐˜’๐˜ฏ๐˜ฐ๐˜ธ๐˜ญ๐˜ฆ๐˜ฅ๐˜จ๐˜ฆ ๐˜ช๐˜ฏ ๐˜Ž๐˜ฆ๐˜ฏ๐˜ฆ๐˜ณ๐˜ข๐˜ต๐˜ช๐˜ท๐˜ฆ ๐˜”๐˜ฐ๐˜ฅ๐˜ฆ๐˜ญ๐˜ด))

35
/ 100
Emerging

This project helps 3D artists, game developers, or computer vision researchers extract intrinsic scene properties like depth, surface normals, albedo, and shading from existing generative models (like Stable Diffusion). You provide a trained generative model and some labeled images, and it outputs detailed intrinsic image components that explain how the generated scene is put together. This is ideal for those who work with synthetic image generation and need to decompose scenes into their fundamental attributes.

191 stars. No commits in the last 6 months.

Use this if you need to understand or modify the underlying physical properties of images generated by AI models, such as separating an object's color from its lighting.

Not ideal if you're not working with generative AI models or if your primary goal is general image editing rather than intrinsic scene decomposition.

3D-rendering game-asset-creation synthetic-data-generation computer-graphics scene-understanding
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

191

Forks

8

Language

Python

License

Last pushed

Oct 21, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/duxiaodan/intrinsic-lora"

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