Pytorch implementation of Conditional image-to-image translation [1] (CVPR 2018)
- Parameters without information in the paper were set arbitrarily. (I could not find the supplementary document)
python train.py --dataset dataset
The following shows basic folder structure.
├── data
├── dataset # not included in this repo
├── trainA
├── aaa.png
├── bbb.jpg
└── ...
├── trainB
├── ccc.png
├── ddd.jpg
└── ...
├── testA
├── eee.png
├── fff.jpg
└── ...
└── testB
├── ggg.png
├── hhh.jpg
└── ...
├── train.py # training code
├── utils.py
├── networks.py
└── name_results # results to be saved here
InputA - InputB - A2B - B2A (this repo) |
- NVIDIA GTX 1080 ti
- cuda 8.0
- python 3.5.3
- pytorch 0.4.0
- torchvision 0.2.1
[1] Lin, Jianxin, et al. "Conditional image-to-image translation." The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)(July 2018). 2018.