Cited from: https://github.com/ali-gtw/ImageStyleTransfer-CNN
This is a PyTorch implementation of Image Style Transfer Using Convolutional Neural Networks, inspired by authors of paper repo.
Coarse-to-fine high-resolution is also added, from paper Controlling Perceptual Factors in Neural Style Transfer.
Simply run python3 main.py.
You may want to change DATA.STYLE_IMG_PATH, DATA.CONTENT_IMG_PATH in config file, for transferring
style of your desired style image to your content image.
Cited from https://github.com/NVIDIA/pix2pixHD
A few changes were made to dataloader and configs for this experiments.
Change the options in ./p2pHD/options to cope with your task. Please refer to the cited repo for more training/testing details.
implementation based on Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
python ./CycleGAN/train.py --options
docker file is provided in ./docker, which provides env to run all the tested models in this repo. For further information regrading docker installation, please refer to docker and nvidia-docker
Please click here to watch the demo video for this project. You can also watch it on YouTube here