This repository is deprecated ! Please refer to our pytorch version Shift-Net_pytorch. That is much faster than this repository. As some code in this repository is implemented using for-loop, while the code of pytorch version Shift-Net_pytorch is fully-implemented parallelly.

If you find this paper useful, please cite:

@InProceedings{Yan_2018_Shift,
author = {Yan, Zhaoyi and Li, Xiaoming and Li, Mu and Zuo, Wangmeng and Shan, Shiguang},
title = {Shift-Net: Image Inpainting via Deep Feature Rearrangement},
booktitle = {The European Conference on Computer Vision (ECCV)},
month = {September},
year = {2018}
}

We benefit a lot from pix2pix and DCGAN. The data loader is modified from pix2pix and the implemetation of Instance Normalization borrows from Instance Normalization. The shift operation is inspired by style-swap.