/ADJSCC

adaptive/attention deep joint source channel coding

Primary LanguagePythonMIT LicenseMIT

Code: Wireless Image Transmission Using Deep Source Channel Coding with Attention Modules

Adaptive/Attention Deep Joint Source Channel Coding Image text

Datasets:

CIFAR-10 is from inner tensorflow.keras.datasets.cifar10.
ImageNet is manually made and is too huge to upload.

Warning:

  1. The provided folder of tensorflow_compression is only for macOS. If you want to use tensorflow_compression in other systems, please use pip to install tensorflow_compression and change corresponding codes reling on tensorflow_compression.
  2. If you want to use ImageNet to test bdjscc_imagenet.py and adjscc_imagenet.py, you can use pip to install tensorflow_dataset and download ImageNet. The corrsponding code of loading ImageNet should be modified.

Citation:

J. Xu, B. Ai, W. Chen, A. Yang, P. Sun and M. Rodrigues, "Wireless Image Transmission Using Deep Source Channel Coding With Attention Modules," in IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 4, pp. 2315-2328, April 2022, doi: 10.1109/TCSVT.2021.3082521.

If you have any question, please feel free to contact me via: xjl-88410@163.com