This is the Pytorch1.4 version of Universal Style Transfer via Feature Transforms.
We translated the original WCT code from Pytorch 0.4.1 to Pytorch 1.4. Details of the code refers to https://github.com/sunshineatnoon/PytorchWCT.
- Pytorch 1.4
- torchvision 0.5.0
- Pretrained encoder and decoder models for image reconstruction only (download and uncompress them under models/). The .t7 files must convert to .pth with Convert code.
- CUDA + CuDNN
Simply put content and image pairs in images/content
and images/style
respectively. Note that correspoding conternt and image pairs should have same names.
python WCT.py --cuda
Li Y, Fang C, Yang J, et al. Universal Style Transfer via Feature Transforms[J]. arXiv preprint arXiv:1705.08086, 2017.