Universal Style Transfer
This is the Pytorch implementation of Universal Style Transfer via Feature Transforms.
Official Torch implementation can be found here and Tensorflow implementation can be found here.
Prerequisites
- Pytorch
- torchvision
- Pretrained encoder and decoder models for image reconstruction only (download and uncompress them under models/)
- CUDA + CuDNN
Prepare images
Simply put content and image pairs in images/content and images/style respectively. Note that correspoding conternt and image pairs should have same names.
Style Transfer
python WCT.py --cuda
Results
Acknowledgments
Many thanks to the author Yijun Li for his kind help.
Reference
Li Y, Fang C, Yang J, et al. Universal Style Transfer via Feature Transforms[J]. arXiv preprint arXiv:1705.08086, 2017.