Copyright@ Yiru Chen

Implement Image-to-Image Translation with Conditional Adversarial Networks on tensorflow.

• Python with numpy
• NVIDIA GPU + CUDA 8.0 + CuDNNv5.1
• TensorFlow 0.12

• Download the dataset (script borrowed from torch code):

bash ./download_dataset.sh facades

• Train the model

bash python train.py

Current code supports CMP Facades dataset. And the latest implementation trains for 200 epochs. You are free to modify the constants in config.py.

Here shows some of the pictures generated from this implementation.

• Datasets: Facades

My implementation referenced pix2pix. Thanks for their excellent work!

Copyright (c) Yiru Chen

Licensed under the Apache License 2.0.