SonyResearch/RAW-Diffusion

[WACV 2025] Official implementation of "RAW-Diffusion: RGB-Guided Diffusion Models for High-Fidelity RAW Image Generation"

32
/ 100
Emerging

This project helps computer vision researchers and data scientists generate realistic, high-fidelity RAW image datasets for specific camera sensors. You provide a collection of standard RGB images and the tool outputs corresponding RAW images, tailored for your desired camera. This is particularly useful for training robust image recognition models, especially for challenging conditions like low light.

No commits in the last 6 months.

Use this if you need to create diverse, camera-specific RAW image datasets for computer vision tasks, even with a limited number of existing RAW images.

Not ideal if you are looking for a general-purpose image generation tool for everyday use or do not specifically work with RAW image data.

computer-vision image-processing dataset-generation machine-learning-research camera-sensor-calibration
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

60

Forks

4

Language

Python

License

Apache-2.0

Last pushed

Feb 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/SonyResearch/RAW-Diffusion"

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