TDCN
This repository is an PyTorch implementation of the paper Tree-structured Dilated Convolutional Networks for Image Compressed Sensing.
You can find the original code and more information from here.
The paper: link
The version of the python package we use is：
python 3.9
torch 1.10.0 + cuda 10.2
torchvision 0.11.0
scipy 1.7.1
glob、pandas and so on
Training
We used DIV2K dataset to train our model. Please download it from here or 百度网盘 提取码：wj46.
Unpack the tar file to any place you want. Then, change the folder argument in generate_train.m to the place where DIV2K images are located. Use train.py for training.
Testing
You can use the model we provide for testing. Use test.py for testing TDCN.
Citation
If you find TDCN useful in your research, please consider citing:
R. Lu and K. Ye, "Tree-Structured Dilated Convolutional Networks for Image Compressed Sensing," in IEEE Access, vol. 10, pp. 98374-98383, 2022, doi: 10.1109/ACCESS.2022.3206016.
If you have any questions, you are welcome to contact me. My email is: 496315130@qq.com.