zsameem / stargan

Official PyTorch Implementation of StarGAN - CVPR 2018

Running in Docker

To build a docker image with the necessary dependencies, dockerfile is provided. Build the image with the command nvidia-docker build -t car2car-pytorch ./

Default command to train on the CompCars dataset is provided in the train_command.sh. To run it inside docker execute the command

nvidia-docker run -ti --volume=$(pwd):<source root dir> \
-w <source root dir> -u $(id -u):$(id -g) car2car-pytorch:latest \ 
./train_command.sh

nvidia-docker run -ti --volume=$(pwd):/localhome/team07/stargan \
-w /localhome/team07/stargan -u $(id -u):$(id -g) car2car-pytorch:latest \
./train_command.sh

Paper

StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation
Yunjey Choi ^1,2, Minje Choi ^1,2, Munyoung Kim ^2,3, Jung-Woo Ha ², Sung Kim ^2,4, and Jaegul Choo ^{1,2    
1} Korea University, ² Clova AI Research (NAVER Corp.), ³ The College of New Jersey, ⁴ HKUST
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018 (Oral)