ianzen/sketchygan

Code for SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis

SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis

Code for "SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis".

Prerequisites

• Python 3, NumPy, SciPy, OpenCV 3
• Tensorflow(>=1.7.0)
• A recent NVIDIA GPU

Preparations

• The path to data files needs to be specified in input_pipeline.py. See below for detailed information on data files.
• You need to download "Inception-V4 model", unzip it and put the checkpoint under inception_v4_model.

Dataset

Pre-built tfrecord files are available for out of the box training.

• Files for the Sketchy Database can be found here.
• Files for Augmented Sketchy(i.e. flickr images+edge maps), resized to 256x256 regardless of original aspect ratios, can be found here.

If you wish to get the image files:

• The Sketchy Database can be found here.
• Use extract_images.py under data_processing to extract images from tfrecord files. You need to specify input and output paths. The extracted images will be sorted by class names.
• Please contact me if you need the original (not resized) Flickr images, since they are too large to upload to any online space.

Configurations

The model can be trained out of the box, by running main_single.py. But there are several places you can change configurations:

• Commandline options in main_single.py
• Some global options in config.py
• Activation/Normalization functions in models_mru.py

Model

• The model will be saved periodically. If you wish to resume, just use commandline switch resume_from.
• If you wish to test the model, change mode from train to test and fill in resume_from.

Citation

If you use our work for your research, please cite our paper

@InProceedings{Chen_2018_CVPR,
author = {Chen, Wengling and Hays, James},
title = {SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2018}
}

Credits

• Incpetion-V4 and VGG16 code by Tensorflow Authors.
• Tensorflow implementation of Spectral Normalization by minhnhat93
• Improved WGAN