https://github.com/zsdonghao/Unsup-Im2Im

Unsupervised Image to Image Translation with Generative Adversarial Networks
https://github.com/zsdonghao/Unsup-Im2Im

deep-learning gan generative-adversarial-network image-to-image tensorflow tensorlayer

Last synced: about 2 months ago
JSON representation

Unsupervised Image to Image Translation with Generative Adversarial Networks

• Host: GitHub
• URL: https://github.com/zsdonghao/Unsup-Im2Im
• Owner: zsdonghao
• Created: 2016-12-17T14:46:12.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2020-07-19T03:13:26.000Z (about 4 years ago)
• Last Synced: 2024-04-11T04:14:05.191Z (5 months ago)
• Topics: deep-learning, gan, generative-adversarial-network, image-to-image, tensorflow, tensorlayer
• Language: Python
• Homepage: https://github.com/zsdonghao/tensorlayer
• Size: 2.11 MB
• Stars: 73
• Watchers: 10
• Forks: 16
• Open Issues: 4
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-tensorflow - im2im - Unsupervised Image to Image Translation with Generative Adversarial Networks (Models/Projects)
• awesome-tensorflow - im2im - Unsupervised Image to Image Translation with Generative Adversarial Networks (Models/Projects)

README

        
## Unsupervised Image to Image Translation with Generative Adversarial Networks





	





##### Paper: [Unsupervised Image to Image Translation with Generative Adversarial Networks](https://arxiv.org/abs/1701.02676)

### Requirements

* TensorFlow 1.0.0

* TensorLayer 1.3.11

* CUDA 8

* Ubuntu 

### Dataset

* Before training the network, please prepare the data

* CelebA [download](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html)

* Cropped SVHN [download](http://ufldl.stanford.edu/housenumbers/)

* MNIST [download](https://github.com/myleott/mnist_png/blob/master/mnist_png.tar.gz), and put to `data/mnist_png`

### Usage

#### Step 1: Learning shared feature

```

python3 train.py --train_step="ac_gan" --retrain=1

```

#### Step 2: Learning image encoder

```

python3 train.py --train_step="imageEncoder" --retrain=1

```

#### Step 3: Translation

```

python3 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`