Awesome Open Source
Awesome Open Source


Implementation of the paper : Toward Multimodal Image-to-Image Translation


Result First column represents input, second column the ground truth. The next is the image generated from cLR-GAN and the last column represents the image generated from cVAE-GAN. Results were obtained from validation dataset.


Model Architecture Visualization

  • Network

Fig 1: Structure of BicycleGAN. (Image taken from the paper)

  • Tensorboard visualization of the entire network

cVAE-GAN Network


  • tensorflow (1.4.0)
  • numpy (1.13.3)
  • scikit-image (0.13.1)
  • scipy (1.0.0)

To install the above dependencies, run:

$ sudo pip install -r requirements.txt




  • Download the datasets from the following links

  • To generate numpy files for the datasets,

    $ python --create <dataset_name>

    This creates train.npy and val.npy in the corresponding dataset directory. This generates very huge files. As an alternate, the next step attempts to read images at run-time during training

  • Alternate to the above step, you could read the images in real time during training. To do this, you should create files containing paths to the images. This can be done by running the following script in the root of this repo.

    $ bash


  • Generating graph:

    To visualize the connections between the graph nodes, we can generate the graph using the flag archi. This would be useful to assert the connections are correct. This generates the graph for bicycleGAN

    $ python --archi

    To generate the model graph for cvae-gan,

    $ python --model cvae-gan --archi

    Possible models are: cvae-gan, clr-gan, bicycle (default)

    To visualize the graph on tensorboard, run the following command:

    $ tensorboard --logdir=logs/summary/Run_1 --host=

    Replace Run_1 with the latest directory name

  • Complete list of options:

    $ python --help

  • Training the network

    To train model (say cvae-gan) on dataset (say facades) from scratch,

    $ python --train --model cvae-gan --dataset facades

    The above command by default trains the model in which images from distribution of domain B are generated conditioned on the images from the distribution of domain A. To switch the direction,

    $ python --train --model cvae-gan --dataset facades --direction b2a

    To resume the training from a checkpoint,

    $ python --resume <path_to_checkpoint> --model cvae-gan

  • Testing the network

    • Download the checkpoint file from here and place the checkpoint files in the ckpt directory

    To test the model from the given trained models, by default the model generates 5 different images (by sampling 5 different noise samples)

    $ ./ <dataset_name> <test_image_path>

    To generate multiple output samples,

    $ ./ <dataset_name> <test_image_path> < # of samples>

    Try it with some of the test samples present in the directory imgs/test


Loss of discriminator and generator as function of iterations on edges2shoes dataset.


  • [x] Residual Encoder
  • [ ] Multiple discriminators for cVAE-GAN and cLR-GAN
  • [ ] Inducing noise to all the layers of the generator
  • [ ] Train the model on rest of the datasets


Released under the MIT license

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
python (50,967
tensorflow (2,100
gans (73
image-generation (63
pix2pix (42
image-translation (28
nips-2017 (18

Find Open Source By Browsing 7,000 Topics Across 59 Categories