[Tensorflow] (by Harry Yang), [Tensorflow] (by Archit Rathore), [Tensorflow] (by Van Huy), [Tensorflow] (by Xiaowei Hu), [Tensorflow-simple] (by Zhenliang He), [TensorLayer] (by luoxier), [Chainer] (by Yanghua Jin), [Minimal PyTorch] (by yunjey), [Mxnet] (by Ldpe2G), [lasagne/Keras] (by tjwei), [Keras] (by Simon Karlsson)

_cpu ``` , where `` can be `horse2zebra`, `style_monet`, etc. You just need to append `_cpu` to the target model. ## Training and Test Details To train a model, ```bash DATA_ROOT=/path/to/data/ name=expt_name th train.lua ``` Models are saved to `./checkpoints/expt_name` (can be changed by passing `checkpoint_dir=your_dir` in train.lua). See `opt_train` in `options.lua` for additional training options. To test the model, ```bash DATA_ROOT=/path/to/data/ name=expt_name phase=test th test.lua ``` This will run the model named `expt_name` in both directions on all images in `/path/to/data/testA` and `/path/to/data/testB`. A webpage with result images will be saved to `./results/expt_name` (can be changed by passing `results_dir=your_dir` in test.lua). See `opt_test` in `options.lua` for additional test options. Please use `model=one_direction_test` if you only would like to generate outputs of the trained network in only one direction, and specify `which_direction=AtoB` or `which_direction=BtoA` to set the direction. There are other options that can be used. For example, you can specify `resize_or_crop=crop` option to avoid resizing the image to squares. This is indeed how we trained GTA2Cityscapes model in the projet [webpage](https://junyanz.github.io/CycleGAN/) and [Cycada](https://arxiv.org/pdf/1711.03213.pdf) model. We prepared the images at 1024px resolution, and used `resize_or_crop=crop fineSize=360` to work with the cropped images of size 360x360. We also used `lambda_identity=1.0`. ## Datasets Download the datasets using the following script. Many of the datasets were collected by other researchers. Please cite their papers if you use the data. ```bash bash ./datasets/download_dataset.sh dataset_name ``` - `facades`: 400 images from the [CMP Facades dataset](http://cmp.felk.cvut.cz/~tylecr1/facade/). [[Citation](datasets/bibtex/facades.tex)] - `cityscapes`: 2975 images from the [Cityscapes training set](https://www.cityscapes-dataset.com/). [[Citation](datasets/bibtex/cityscapes.tex)]. Note: Due to license issue, we do not host the dataset on our repo. Please download the dataset directly from the Cityscapes webpage. Please refer to `./datasets/prepare_cityscapes_dataset.py` for more detail. - `maps`: 1096 training images scraped from Google Maps. - `horse2zebra`: 939 horse images and 1177 zebra images downloaded from [ImageNet](http://www.image-net.org/) using the keywords `wild horse` and `zebra` - `apple2orange`: 996 apple images and 1020 orange images downloaded from [ImageNet](http://www.image-net.org/) using the keywords `apple` and `navel orange`. - `summer2winter_yosemite`: 1273 summer Yosemite images and 854 winter Yosemite images were downloaded using Flickr API. See more details in our paper. - `monet2photo`, `vangogh2photo`, `ukiyoe2photo`, `cezanne2photo`: The art images were downloaded from [Wikiart](https://www.wikiart.org/). The real photos are downloaded from Flickr using the combination of the tags *landscape* and *landscapephotography*. The training set size of each class is Monet:1074, Cezanne:584, Van Gogh:401, Ukiyo-e:1433, Photographs:6853. - `iphone2dslr_flower`: both classes of images were downloaded from Flickr. The training set size of each class is iPhone:1813, DSLR:3316. See more details in our paper. ## Display UI Optionally, for displaying images during training and test, use the [display package](https://github.com/szym/display). - Install it with: `luarocks install https://raw.githubusercontent.com/szym/display/master/display-scm-0.rockspec` - Then start the server with: `th -ldisplay.start` - Open this URL in your browser: [http://localhost:8000](http://localhost:8000) By default, the server listens on localhost. Pass `0.0.0.0` to allow external connections on any interface: ```bash th -ldisplay.start 8000 0.0.0.0 ``` Then open `http://(hostname):(port)/` in your browser to load the remote desktop. ## Setup Training and Test data To train CycleGAN model on your own datasets, you need to create a data folder with two subdirectories `trainA` and `trainB` that contain images from domain A and B. You can test your model on your training set by setting ``phase='train'`` in `test.lua`. You can also create subdirectories `testA` and `testB` if you have test data. You should **not** expect our method to work on just any random combination of input and output datasets (e.g. `cats<->keyboards`). From our experiments, we find it works better if two datasets share similar visual content. For example, `landscape painting<->landscape photographs` works much better than `portrait painting <-> landscape photographs`. `zebras<->horses` achieves compelling results while `cats<->dogs` completely fails. See the following section for more discussion. ## Failure cases Our model does not work well when the test image is rather different from the images on which the model is trained, as is the case in the figure to the left (we trained on horses and zebras without riders, but test here one a horse with a rider). See additional typical failure cases [here](https://junyanz.github.io/CycleGAN/images/failures.jpg). On translation tasks that involve color and texture changes, like many of those reported above, the method often succeeds. We have also explored tasks that require geometric changes, with little success. For example, on the task of `dog<->cat` transfiguration, the learned translation degenerates into making minimal changes to the input. We also observe a lingering gap between the results achievable with paired training data and those achieved by our unpaired method. In some cases, this gap may be very hard -- or even impossible,-- to close: for example, our method sometimes permutes the labels for tree and building in the output of the cityscapes photos->labels task. ## Citation If you use this code for your research, please cite our [paper](https://junyanz.github.io/CycleGAN/): ``` @inproceedings{CycleGAN2017, title={Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkss}, author={Zhu, Jun-Yan and Park, Taesung and Isola, Phillip and Efros, Alexei A}, booktitle={Computer Vision (ICCV), 2017 IEEE International Conference on}, year={2017} } ``` ## Related Projects: **[contrastive-unpaired-translation](https://github.com/taesungp/contrastive-unpaired-translation) (CUT)**
**[pix2pix-Torch](https://github.com/phillipi/pix2pix) | [pix2pixHD](https://github.com/NVIDIA/pix2pixHD) | [BicycleGAN](https://github.com/junyanz/BicycleGAN) | [vid2vid](https://tcwang0509.github.io/vid2vid/) | [SPADE/GauGAN](https://github.com/NVlabs/SPADE)**
**[iGAN](https://github.com/junyanz/iGAN) | [GAN Dissection](https://github.com/CSAILVision/GANDissect) | [GAN Paint](http://ganpaint.io/)** ## Cat Paper Collection If you love cats, and love reading cool graphics, vision, and ML papers, please check out the Cat Paper [Collection](https://github.com/junyanz/CatPapers). ## Acknowledgments Code borrows from [pix2pix](https://github.com/phillipi/pix2pix) and [DCGAN](https://github.com/soumith/dcgan.torch). The data loader is modified from [DCGAN](https://github.com/soumith/dcgan.torch) and [Context-Encoder](https://github.com/pathak22/context-encoder). The generative network is adopted from [neural-style](https://github.com/jcjohnson/neural-style) with [Instance Normalization](https://github.com/DmitryUlyanov/texture_nets/blob/master/InstanceNormalization.lua).