Skip to content
TensorFlow Implementation of Unsupervised Cross-Domain Image Generation
Python Shell
Branch: master
Clone or download
Latest commit eaa2d5e Jan 25, 2017
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
jpg Add files via upload Jan 21, 2017
LICENSE Initial commit Dec 30, 2016
README.md Update README.md Jan 21, 2017
download.sh download svhn dataset Jan 21, 2017
main.py add code for creating directories Jan 22, 2017
model.py name scope added Jan 24, 2017
prepro.py download mnist and resize to 32x32 Jan 21, 2017
solver.py train and eval the model Jan 22, 2017

README.md

Domain Transfer Network (DTN)

TensorFlow implementation of Unsupervised Cross-Domain Image Generation. alt text

Requirements


Usage

Clone the repository

$ git clone https://github.com/yunjey/dtn-tensorflow.git
$ cd dtn-tensorflow

Download the dataset

$ chmod +x download.sh
$ ./download.sh

Resize MNIST dataset to 32x32

$ python prepro.py

Pretrain the model f

$ python main.py --mode='pretrain'

Train the model G and D

$ python main.py --mode='train'

Transfer SVHN to MNIST

$ python main.py --mode='eval'

Results

From SVHN to MNIST

alt text

alt text

alt text

alt text

From Photos to Emoji (in paper)

alt text

alt text

You can’t perform that action at this time.