ViSTRA2: Video coding using spatial resolution and effective bit depth adaptation

作者：

Highlights：

• A new video compression framework (ViSTRA2) which exploits adaptation of spatial resolution and effective bit depth.

• ViSTRA2 down-samples these parameters at the encoder based on perceptual criteria, and upsampling at the decoder using a deep convolution neural network.

• It has been integrated with HEVC and VVC and show consistent and significant coding gains.

摘要

•A new video compression framework (ViSTRA2) which exploits adaptation of spatial resolution and effective bit depth.•ViSTRA2 down-samples these parameters at the encoder based on perceptual criteria, and upsampling at the decoder using a deep convolution neural network.•It has been integrated with HEVC and VVC and show consistent and significant coding gains.

论文关键词：Spatial resolution adaptation,Effective bit depth adaptation,Video compression,Machine learning based compression,HEVC,VVC

论文评审过程：Received 25 March 2020, Revised 26 May 2021, Accepted 9 June 2021, Available online 17 June 2021, Version of Record 19 June 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116355