mlsanityData to 7 Oct 2026

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

arXiv 2203.16357 ↗First posted 30 Mar 2022eess.IV cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 28 of its 46 references that are in the map. First posted on arXiv 30 Mar 2022.

Builds on

28 of 46 references.

References · 28

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  3. 2018Variational image compression with a scale hyperpriorcited by 910
  4. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  5. 2016Pixel Recurrent Neural Networkscited by 814
  6. 2016End-to-end Optimized Image Compressioncited by 673
  7. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 642
  8. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 570
  9. 2020Channel-wise Autoregressive Entropy Models for Learned Image Compressioncited by 255
  10. 2021Checkerboard Context Model for Efficient Learned Image Compressioncited by 193
  11. 2018Context-adaptive Entropy Model for End-to-end Optimized Image Compressioncited by 138
  12. 2017Real-Time Adaptive Image Compressioncited by 127
16 more
  1. 2018Practical Full Resolution Learned Lossless Image Compressioncited by 93
  2. 2019Non-local Attention Optimized Deep Image Compressioncited by 86
  3. 2019Practical Lossless Compression with Latent Variables using Bits Back Codingcited by 85
  4. 2019Integer Discrete Flows and Lossless Compressioncited by 82
  5. 2017Parallel Multiscale Autoregressive Density Estimationcited by 71
  6. 2020Causal Contextual Prediction for Learned Image Compressioncited by 70
  7. 2020Quantization Guided JPEG Artifact Correctioncited by 57
  8. 2019Bit-Swap: Recursive Bits-Back Coding for Lossless Compression with Hierarchical Latent Variablescited by 51
  9. 2019HiLLoC: Lossless Image Compression with Hierarchical Latent Variable Modelscited by 36
  10. 2020IDF++: Analyzing and Improving Integer Discrete Flows for Lossless Compressioncited by 35
  11. 2019Computationally Efficient Neural Image Compressioncited by 31
  12. 2021iVPF: Numerical Invertible Volume Preserving Flow for Efficient Lossless Compressioncited by 27
  13. 2020Lossless Image Compression through Super-Resolutioncited by 25
  14. 2017The Design, Implementation, and Deployment of a System to Transparently Compress Hundreds of Petabytes of Image Files for a File-Storage Servicecited by 6
  15. 2021Transform Network Architectures for Deep Learning based End-to-End Image/Video Coding in Subsampled Color Spacescited by 6
  16. 2021Lossless Image Compression Using a Multi-Scale Progressive Statistical Modelcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 0; 2025: 1; 2026: 1.

4 papers.

  1. 2023Make Lossy Compression Meaningful for Low-Light Imagescited by 3
  2. 2023Unified learning-based lossy and lossless JPEG recompressioncited by 0
  3. 2025JPEG Processing Neural Operator for Backward-Compatible Codingcited by 0
  4. 2026FLM: Frequency-Aware Language Models for Generative Image Compressioncited by 0