GitHub - wzhshi/TIP-CSNet: The training codes, the training data, and some pre-trained models for my TIP paper "Image Compressed Sensing using Convolutional Neural Network".

The test code for my TIP paper "Image Compressed Sensing using Convolutional Neural Network" will be released here. Please look forward to it.

%-----------------2019.08.17--------------------------------

The training codes, the training data, and some pre-trained models are given. Please feel free to contact me if you have any questions. My email is wzhshi@hit.edu.cn

To run the code, you should copy the file "bcs_init_rec_dag.m" in the folder "myLayers" into the folder "+dagnn" of your Matconvnet.

If the code is useful to you, please cite our papers:

@Article{shi2019image,

title={Image compressed sensing using convolutional neural network},

author={Shi, Wuzhen and Jiang, Feng and Liu, shaohui and Zhao, Debin},

journal={IEEE Transactions on Image Processing},

volume = {29},

pages = {375--388},

year={2020}

}

@inproceedings{shi2017deep,

title={Deep networks for compressed image sensing},

author={Shi, Wuzhen and Jiang, Feng and Zhang, Shengping and Zhao, Debin},

booktitle={Multimedia and Expo (ICME), 2017 IEEE International Conference on},

pages={877--882},

year={2017},

organization={IEEE}

}