JunlinHan/DCLGAN

Code for Dual Contrastive Learning for Unsupervised Image-to-Image Translation, NTIRE, CVPRW 2021, oral.

45
/ 100
Emerging

This tool helps researchers and visual effects artists transform images from one style to another without needing matching pairs of images. For instance, you can take a collection of photos of horses and a collection of photos of zebras, and the tool will generate realistic 'zebra' versions of your horses and 'horse' versions of your zebras. It's ideal for anyone working with visual data who needs to bridge visual gaps between distinct image domains.

178 stars. No commits in the last 6 months.

Use this if you need to convert a collection of images from one visual domain to another, like changing seasons in photos or converting sketches to photos, without having precisely paired examples for training.

Not ideal if your image translation task requires highly precise, pixel-perfect transformations where slight visual artifacts are unacceptable, or if you need to work with very small, unbalanced datasets without dedicated mode collapse solutions.

image-transformation visual-effects computer-vision digital-art data-augmentation
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

178

Forks

24

Language

Python

License

Last pushed

Jul 10, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/JunlinHan/DCLGAN"

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