Unsupervised Image to Image Translation with Generative Adversarial Networks
Switch branches/tags
Nothing to show
Clone or download
Latest commit ef5d029 Nov 23, 2017
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
img remove ._ Apr 24, 2017
README.md Update README.md Nov 22, 2017
data_loader.py release TF1.0 Apr 24, 2017
model.py release TF1.0 Apr 24, 2017
train.py Update train.py Jun 11, 2017
translate_image.py release TF1.0 Apr 24, 2017
utils.py release TF1.0 Apr 24, 2017

README.md

Unsupervised Image to Image Translation with Generative Adversarial Networks

Paper: Unsupervised Image to Image Translation with Generative Adversarial Networks

Dataset

  • Before training the network, please prepare the data
  • CelebA download
  • SVHN download
  • MNIST download, and put to data/mnist_png

Usage

Step 1: Learning shared feature

python train.py --train_step="ac_gan" --retrain=1

Step 2: Learning image encoder

python train.py --train_step="imageEncoder" --retrain=1

Step 3: Translation

python translate_image.py
  • Samples of all steps will be saved to data/samples/

Network

Want to use different datasets?

  • in train.py and translate_image.py modify the name of dataset flags.DEFINE_string("dataset", "celebA", "The name of dataset [celebA, obama_hillary]")
  • write your own data_loader in data_loader.py