kmsiapps / e2e-dnn-comm-for-image

End to end deep neural network based semantic communication system for image.
64 stars 6 forks source link

End to end deep neural network based semantic communication system for image

Code repository for ICC 2022 demo proposal "Real-Time Semantic Communications with a Vision Transformer".

You may want to see On the Role of ViT and CNN in Semantic Communications: Analysis and Prototype Validation, a successor of this project.

Requirements

tensorflow tensorflow_addons

Get started

python ./train.py for training python ./evalute.py for evalutation

You may refer to this issue for further instructions. There is currently no command line arguments; please edit the codes before you run them.

System architecture

sysarch

Results

results-summary

Reference

@INPROCEEDINGS{yoo2022real,
  author={Yoo, Hanju and Jung, Taehun and Dai, Linglong and Kim, Songkuk and Chae, Chan-Byoung},
  booktitle={2022 IEEE International Conference on Communications Workshops (ICC Workshops)}, 
  title={Demo: Real-Time Semantic Communications with a Vision Transformer}, 
  year={2022},
  volume={},
  number={},
  pages={1-2},
  doi={10.1109/ICCWorkshops53468.2022.9914635}}