Efficient and Effective Context-Based Convolutional Entropy Modelling for Image Compression

Published in IEEE Transactions on Image Processing (TIP), 2020

This paper propose a group context which enables parallel encoding and decoding inner groups for the 3D code block produced in DNN-based compression framework.

Demo

Download paper here

Recommended citation:

@ARTICLE{li2020efficient,
 author={Li, Mu and Ma, Kede and You, Jane and Zhang, David and Zuo, Wangmeng},
 journal={IEEE Transactions on Image Processing},
 title={Efficient and Effective Context-Based Convolutional Entropy Modeling for Image Compression},
 year={2020},
 volume={29},
 pages={5900-5911},
 doi={10.1109/TIP.2020.2985225}
}