ViSTRA3: Video Coding with Deep Parameter Adaptation and Post Processing

Chen Feng

Duolikun Danier

Charlie Tan

Fan Zhang

David Bull

University of Bristol

About

This paper presents a deep learning-based video compression framework (ViSTRA3). The proposed framework intelligently adapts video format parameters of the input video before encoding, subsequently employing a CNN at the decoder to restore their original format and enhance reconstruction quality. ViSTRA3 has been integrated with the H.266/VVC Test Model VTM 14.0, and evaluated under the Joint Video Exploration Team Common Test Conditions. Bjønegaard Delta (BD) measurement results show that the proposed framework consistently outperforms the original VVC VTM, with average BD-rate savings of 1.8% and 3.7% based on the assessment of PSNR and VMAF.

Source code

The source code of MFRNet has been released at here

Framework and Model

Results

Participated the Challenge on Learned Image Compression (CLIC) in IEEE/CVF CVPR 2022, and ranks top six in the video track..

Citation

@INPROCEEDINGS{9937265,
  author={Feng, Chen and Danier, Duolikun and Tan, Charlie and Zhang, Fan and Bull, David},
  booktitle={2022 IEEE International Symposium on Circuits and Systems (ISCAS)}, 
  title={ViSTRA3: Video Coding with Deep Parameter Adaptation and Post Processing}, 
  year={2022},
  volume={},
  number={},
  pages={824-828},
  doi={10.1109/ISCAS48785.2022.9937265}}[paper]