Unofficical Pytorch implementation of the papaer Semantically Multi-modal Image Synthesis - Zhen Zhu, Zhiliang Xu, Ansheng You, and Xiang Bai in CVPR 2020.
Official impelemntation will be released soon.
- python 3.7
- pytorch 1.4.0
python3 -m pip install -r requirements.txt
Download Cityscapes Dataset from (https://www.cityscapes-dataset.com/)
python3 main_train.py --data_root ~/dataset/cityscapes --batch_size 1
Save output images to directory checkpoint/test/