mlsanityData to 14 Oct 2026

Rethinking Learned Image Compression: Context is All You Need

arXiv 2407.11590 ↗First posted 16 Jul 2024eess.IV cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

Since LIC has made rapid progress recently compared to traditional methods, this paper attempts to discuss the question about 'Where is the boundary of Learned Image Compression(LIC)?'. Thus this paper splits the above problem into two sub-problems:1)Where is the boundary of rate-distortion performance of PSNR? 2)How to further improve the compression gain and achieve the boundary? Therefore this paper analyzes the effectiveness of scaling parameters for encoder, decoder and context model, which are the three components of LIC. Then we conclude that scaling for LIC is to scale for context model and decoder within LIC. Extensive experiments demonstrate that overfitting can actually serve as an effective context. By optimizing the context, this paper further improves PSNR and achieves state-of-the-art performance, showing a performance gain of 14.39% with BD-RATE over VVC.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 52 of its 82 references that are in the map. First posted on arXiv 16 Jul 2024.

Builds on

52 of 82 references.

References · 52

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2022Training Compute-Optimal Large Language Modelscited by 3,019
  5. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  6. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
  7. 2020Scaling Laws for Autoregressive Generative Modelingcited by 522
  8. 2020High-Fidelity Generative Image Compressioncited by 338
  9. 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 301
  10. 2020Channel-wise Autoregressive Entropy Models for Learned Image Compressioncited by 264
  11. 2018Generative Adversarial Networks for Extreme Learned Image Compressioncited by 219
  12. 2019Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoffcited by 211
38 more
  1. 2023Learned Image Compression with Mixed Transformer-CNN Architecturescited by 211
  2. 2021Checkerboard Context Model for Efficient Learned Image Compressioncited by 202
  3. 2016Full Resolution Image Compression with Recurrent Neural Networkscited by 175
  4. 2015Variable Rate Image Compression with Recurrent Neural Networkscited by 173
  5. 2018Context-adaptive Entropy Model for End-to-end Optimized Image Compressioncited by 143
  6. 2024Neural Video Compression with Feature Modulationcited by 123
  7. 2019Variable Rate Deep Image Compression With a Conditional Autoencodercited by 114
  8. 2022Broken Neural Scaling Lawscited by 109
  9. 2020Video Coding for Machines: A Paradigm of Collaborative Compression and Intelligent Analyticscited by 80
  10. 2022Multi-Realism Image Compression with a Conditional Generatorcited by 80
  11. 2021Transformer-based Image Compressioncited by 62
  12. 2021Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transformcited by 58
  13. 2023Frequency-Aware Transformer for Learned Image Compressioncited by 58
  14. 2021Soft then Hard: Rethinking the Quantization in Neural Image Compressioncited by 51
  15. 2018Deep feature compression for collaborative object detectioncited by 48
  16. 2021Image coding for machines: an end-to-end learned approachcited by 47
  17. 2022PO-ELIC: Perception-Oriented Efficient Learned Image Codingcited by 36
  18. 2019Efficient and Effective Context-Based Convolutional Entropy Modeling for Image Compressioncited by 33
  19. 2020Towards Coding for Human and Machine Vision: A Scalable Image Coding Approachcited by 31
  20. 2020End-to-end optimized image compression for machines, a studycited by 30
  21. 2021Learned Image Coding for Machines: A Content-Adaptive Approachcited by 21
  22. 2023LVQAC: Lattice Vector Quantization Coupled with Spatially Adaptive Companding for Efficient Learned Image Compressioncited by 19
  23. 2022Selective compression learning of latent representations for variable-rate image compressioncited by 17
  24. 2023MobileNVC: Real-time 1080p Neural Video Compression on a Mobile Devicecited by 13
  25. 2021Interpolation variable rate image compressioncited by 12
  26. 2023Semantically Structured Image Compression via Irregular Group-Based Decouplingcited by 12
  27. 2020End-to-end Learned Image Compression with Fixed Point Weight Quantizationcited by 10
  28. 2022Q-LIC: Quantizing Learned Image Compression with Channel Splittingcited by 10
  29. 2022Rate-Distortion Optimized Post-Training Quantization for Learned Image Compressioncited by 10
  30. 2021Image Compression with Recurrent Neural Network and Generalized Divisive Normalizationcited by 9
  31. 2023Transformer-based Variable-rate Image Compression with Region-of-interest Controlcited by 7
  32. 2020Channel-Level Variable Quantization Network for Deep Image Compressioncited by 6
  33. 2022Identity Preserving Loss for Learned Image Compressioncited by 4
  34. 2024Super-High-Fidelity Image Compression via Hierarchical-ROI and Adaptive Quantizationcited by 4
  35. 2021Online Meta Adaptation for Variable-Rate Learned Image Compressioncited by 3
  36. 2022Evaluating the Practicality of Learned Image Compressioncited by 3
  37. 2022Recognition-Aware Learned Image Compressioncited by 2
  38. 2024End-to-End Optimized Image Compression with the Frequency-Oriented Transformcited by 2

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Pad'e Neurons for Efficient Neural Modelscited by 0