Practical Learned Lossless JPEG Recompression with Multi-Level Cross-Channel Entropy Model in the DCT Domain

TL;DR

This paper introduces a deep learning-based method for lossless JPEG recompression in the DCT domain, utilizing a multi-level cross-channel entropy model to achieve state-of-the-art results and surpass traditional methods.

Contribution

It presents the first learned JPEG recompression technique that operates losslessly in the DCT domain, improving storage efficiency over existing traditional methods.

Findings

Achieves state-of-the-art recompression performance

Outperforms traditional methods like Lepton, JPEG XL, and CMIX

First learned method for lossless JPEG transcoding

Abstract

JPEG is a popular image compression method widely used by individuals, data center, cloud storage and network filesystems. However, most recent progress on image compression mainly focuses on uncompressed images while ignoring trillions of already-existing JPEG images. To compress these JPEG images adequately and restore them back to JPEG format losslessly when needed, we propose a deep learning based JPEG recompression method that operates on DCT domain and propose a Multi-Level Cross-Channel Entropy Model to compress the most informative Y component. Experiments show that our method achieves state-of-the-art performance compared with traditional JPEG recompression methods including Lepton, JPEG XL and CMIX. To the best of our knowledge, this is the first learned compression method that losslessly transcodes JPEG images to more storage-saving bitstreams.