blueGorae/DynaGAN
DynaGAN: Dynamic Few-shot Adaptation of GANs to Multiple Domains (SIGGRAPH Asia 2022)
This tool helps researchers and artists adapt existing generative AI models (GANs) to create new, unique image sets with very little new training data. You provide a few example images for several new visual styles or categories, and it outputs a single, dynamically adaptable model that can generate high-quality images across all those styles. This is ideal for those working in AI research, digital content creation, or visual effects who need to quickly expand a model's capabilities without extensive retraining.
No commits in the last 6 months.
Use this if you need to generate images across multiple distinct visual styles or domains, but only have a small number of example images for each new style.
Not ideal if you are looking for a tool to train a generative model from scratch or if you only need to adapt a model to a single new domain.
Stars
45
Forks
9
Language
Python
License
MIT
Category
Last pushed
Jan 10, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/blueGorae/DynaGAN"
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