rish-16/CycleGANsformer
Unpaired Image-to-Image Translation with Transformer-based GANs in PyTorch [WIP]
This tool helps researchers and deep learning practitioners experiment with converting images from one visual domain to another, even when there are no direct pairs of 'before' and 'after' images. For example, you can transform photos of horses into zebras, or turn a landscape photo into a painting. You provide two collections of images (e.g., many horse photos and many zebra photos), and it learns to translate between them, producing new images in the target style or domain.
No commits in the last 6 months.
Use this if you are a researcher or advanced practitioner exploring novel methods for unpaired image-to-image translation, particularly with transformer-based GANs.
Not ideal if you need a production-ready solution for general image manipulation, require precise control over specific object transformations, or are not comfortable with deep learning model training.
Stars
65
Forks
9
Language
Python
License
MIT
Category
Last pushed
Jul 29, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/rish-16/CycleGANsformer"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
yunjey/domain-transfer-network
TensorFlow Implementation of Unsupervised Cross-Domain Image Generation
taesungp/contrastive-unpaired-translation
Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV...
PaddlePaddle/PaddleGAN
PaddlePaddle GAN library, including lots of interesting applications like First-Order motion...
tohinz/ConSinGAN
PyTorch implementation of "Improved Techniques for Training Single-Image GANs" (WACV-21)
sagiebenaim/DistanceGAN
Pytorch implementation of "One-Sided Unsupervised Domain Mapping" NIPS 2017