taesungp/contrastive-unpaired-translation

Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV 2020, in PyTorch)

50
/ 100
Established

This project helps artists, designers, or marketers transform the style or characteristics of one set of images to another, without needing perfectly matched examples. For instance, you could take a collection of photos of Russian Blue cats and generate new images of them as 'Grumpy Cats'. You input two distinct collections of images, and it outputs new images that blend content from one with the style of the other.

2,491 stars. No commits in the last 6 months.

Use this if you need to translate images from one visual domain to another, like changing seasons in a landscape or altering an animal's breed, and only have unpaired collections of source and target images.

Not ideal if you have precisely matched pairs of 'before' and 'after' images for your translation task, as simpler methods might be more suitable and easier to control.

image-style-transfer visual-content-creation digital-art photo-manipulation marketing-visuals
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 24 / 25

How are scores calculated?

Stars

2,491

Forks

446

Language

Python

License

Last pushed

Sep 05, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/taesungp/contrastive-unpaired-translation"

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