README.md

Improved Nonlinear Transform Source-Channel Coding to Catalyze Semantic Communications [pdf]

This is the repository of the paper "Improved Nonlinear Transform Source-Channel Coding to Catalyze Semantic Communications".

Pipeline

Evaluation on Kodak Dataset

Evaluation on CLIC 2021 testset

Requirements

Clone the repo and create a conda environment (we use PyTorch 1.9, CUDA 11.1).

The dependencies includes CompressAI, Natten, and timm.

Trained Models

Download the pre-trained models from Google Drive.

Note: We reorganize code and the performances are slightly different from the paper's.

Evaluate

Evaluate:

# Kodak
python main.py --gpu-id 0 --test-only --eval-dataset-path /path/to/kodak --eval-dataset-name kodak --pretrained /path/to/checkpoint

TODO:

Release the code for online inference (NTSCC++)

Acknowledgement

Codebase from CompressAI, TinyLIC, and Swin Transformer

Citation

If you find this code useful for your research, please cite our paper

@inproceedings{
      wang2023improved,
      title={Improved Nonlinear Transform Source-Channel Coding to Catalyze Semantic Communications},
      author={Sixian Wang and Jincheng Dai and Xiaoqi Qin and Zhongwei Si and Kai Niu and Ping Zhang},
      year={2023},
      booktitle={IEEE Journal of Selected Topics in Signal Processing, early access},
}