mtkresearch/shortest-path-diffusion

Official code for the paper "Image generation with shortest path diffusion" accepted at ICML 2023.

37
/ 100
Emerging

This project helps researchers and practitioners in generative AI create high-quality synthetic images. It takes an existing image and corrupts it using a novel "shortest path" method, then learns to reverse this process to generate new images that are sharper and more realistic than previous approaches. It is intended for those working on advanced image synthesis and computational photography.

No commits in the last 6 months.

Use this if you are an AI researcher or a developer focusing on image generation and want to produce higher fidelity synthetic images using an optimized diffusion model.

Not ideal if you are looking for a plug-and-play image generation tool without needing to engage with model training or command-line interfaces.

generative-ai image-synthesis computational-photography machine-learning-research diffusion-models
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

24

Forks

5

Language

Python

License

Apache-2.0

Last pushed

Jul 10, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/mtkresearch/shortest-path-diffusion"

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