Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/carpedm20/discogan-pytorch
PyTorch implementation of "Learning to Discover Cross-Domain Relations with Generative Adversarial Networks"
https://github.com/carpedm20/discogan-pytorch
gan generative-model pytorch unsupervised-learning
Last synced: 6 days ago
JSON representation
PyTorch implementation of "Learning to Discover Cross-Domain Relations with Generative Adversarial Networks"
- Host: GitHub
- URL: https://github.com/carpedm20/discogan-pytorch
- Owner: carpedm20
- License: apache-2.0
- Created: 2017-03-17T10:26:12.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-03-26T15:26:18.000Z (over 6 years ago)
- Last Synced: 2024-08-08T23:19:42.124Z (3 months ago)
- Topics: gan, generative-model, pytorch, unsupervised-learning
- Language: Jupyter Notebook
- Homepage:
- Size: 79.9 MB
- Stars: 1,085
- Watchers: 40
- Forks: 224
- Open Issues: 12
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# DiscoGAN in PyTorch
PyTorch implementation of [Learning to Discover Cross-Domain Relations with Generative Adversarial Networks](https://arxiv.org/abs/1703.05192).
**\* All samples in README.md are genearted by neural network except the first image for each row.**
\* Network structure is slightly diffferent ([here](https://github.com/carpedm20/DiscoGAN-pytorch/blob/master/models.py#L13-L32)) from the author's [code](https://github.com/SKTBrain/DiscoGAN/blob/master/discogan/model.py#L69-L125).
## Requirements
- Python 2.7
- [Pillow](https://pillow.readthedocs.io/en/4.0.x/)
- [tqdm](https://github.com/tqdm/tqdm)
- [PyTorch](https://github.com/pytorch/pytorch)
- [torch-vision](https://github.com/pytorch/vision)
## Usage
First download datasets (from [pix2pix](https://github.com/phillipi/pix2pix)) with:
$ bash ./data/download_dataset.sh dataset_name
- `facades`: 400 images from [CMP Facades dataset](http://cmp.felk.cvut.cz/~tylecr1/facade/).
- `cityscapes`: 2975 images from the [Cityscapes training set](https://www.cityscapes-dataset.com/).
- `maps`: 1096 training images scraped from Google Maps
- `edges2shoes`: 50k training images from [UT Zappos50K dataset](http://vision.cs.utexas.edu/projects/finegrained/utzap50k/).
- `edges2handbags`: 137K Amazon Handbag images from [iGAN project](https://github.com/junyanz/iGAN).
or you can use your own dataset by placing images like:
data
├── YOUR_DATASET_NAME
│ ├── A
│ | ├── xxx.jpg (name doesn't matter)
│ | ├── yyy.jpg
│ | └── ...
│ └── B
│ ├── zzz.jpg
│ ├── www.jpg
│ └── ...
└── download_dataset.sh
**All images in each dataset should have same size** like using [imagemagick](https://www.imagemagick.org/script/index.php):
# for Ubuntu
$ sudo apt-get install imagemagick
$ mogrify -resize 256x256! -quality 100 -path YOUR_DATASET_NAME/A/*.jpg
$ mogrify -resize 256x256! -quality 100 -path YOUR_DATASET_NAME/B/*.jpg
# for Mac
$ brew install imagemagick
$ mogrify -resize 256x256! -quality 100 -path YOUR_DATASET_NAME/A/*.jpg
$ mogrify -resize 256x256! -quality 100 -path YOUR_DATASET_NAME/B/*.jpg
# for scale and center crop
$ mogrify -resize 256x256^ -gravity center -crop 256x256+0+0 -quality 100 -path ../A/*.jpg
To train a model:
$ python main.py --dataset=edges2shoes --num_gpu=1
$ python main.py --dataset=YOUR_DATASET_NAME --num_gpu=4
To test a model (use your `load_path`):
$ python main.py --dataset=edges2handbags --load_path=logs/edges2handbags_2017-03-18_10-55-37 --num_gpu=0 --is_train=False
## Results
### 1. Toy dataset
Result of samples from 2-dimensional Gaussian mixture models. [IPython notebook](./notebooks/DiscoGAN.ipynb)
**# iteration: 0**:
**# iteration: 10000**:
### 2. Shoes2handbags dataset
**# iteration: 11200**:
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (shoe -> handbag -> shoe)
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (handbag -> shoe -> handbag)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` -> `G_AB(G_BA(G_AB(x_A)))` -> `G_BA(G_AB(G_BA(G_AB(x_A))))` -> ...
### 3. Edges2shoes dataset
**# iteration: 9600**:
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (color -> sketch -> color)
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (sketch -> color -> sketch)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` -> `G_AB(G_BA(G_AB(x_A)))` -> `G_BA(G_AB(G_BA(G_AB(x_A))))` -> ...
### 4. Edges2handbags dataset
**# iteration: 9500**:
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (color -> sketch -> color)
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (sketch -> color -> sketch)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` -> `G_AB(G_BA(G_AB(x_A)))` -> `G_BA(G_AB(G_BA(G_AB(x_A))))` -> ...
### 5. Cityscapes dataset
**# iteration: 8350**:
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (image -> segmentation -> image)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (segmentation -> image -> segmentation)
### 6. Map dataset
**# iteration: 22200**:
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (image -> segmentation -> image)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (segmentation -> image -> segmentation)
### 7. Facades dataset
Generation and reconstruction on dense segmentation dataset looks weird which are not included in the paper.
I guess a naive choice of `mean square error` loss for reconstruction need some change on this dataset.
**# iteration: 19450**:
`x_B` -> `G_BA(x_B)` -> `G_AB(G_BA(x_B))` (image -> segmentation -> image)
`x_A` -> `G_AB(x_A)` -> `G_BA(G_AB(x_A))` (segmentation -> image -> segmentation)
## Related works
- [DCGAN-tensorflow](https://github.com/carpedm20/DCGAN-tensorflow)
- [BEGAN-tensorflow](https://github.com/carpedm20/BEGAN-tensorflow)
- [simulated-unsupervised-tensorflow](https://github.com/carpedm20/simulated-unsupervised-tensorflow)
## Author
Taehoon Kim / [@carpedm20](http://carpedm20.github.io)