In this repository we provide the official implementation of DeepMMSE.

• Codename: DeepMMSE (TSP 2022)

• Code Writers: Peikang Lin ([email protected]);

• Authors: Mingqin Chen ([email protected]); Peikang Lin ([email protected]); Yuhui Quan ([email protected]); Tongyao Pang ([email protected]); Hui Ji ([email protected])

• Institute: School of Computer Science and Engineering, South China University of Technology; Department of Mathematics, National University of Singapore

For more information please see:

• [paper]
• [supmat]
• [website]

Here is the list of libraries you need to install to execute the code:

• Python 3.6
• PyTorch 1.7
• scikit-image
• scipy
• cv2 (opencv for python)

All of them can be installed via conda (anaconda), e.g.

conda install scikit-image

Run the demo code in train.py.

@article{chen2022unsupervised,
author={Chen, Mingqin and Lin, Peikang and Quan, Yuhui and Pang, Tongyao and Ji, Hui},
journal={IEEE Transactions on Signal Processing}, 
title={Unsupervised Phase Retrieval Using Deep Approximate MMSE Estimation}, 
year={2022},
volume={70},
pages={2239-2252}
}

For questions, please send an email to [email protected] or [email protected]

This code is under clearing up. The final version will be released in https://github.com/scut-mingqinchen soon.