mlsanityData to 14 Oct 2026

Synonymous Variational Inference for Perceptual Image Compression

arXiv 2505.22438 ↗First posted 28 May 2025cs.IT cs.AI cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Recent contributions of semantic information theory reveal the set-element relationship between semantic and syntactic information, represented as synonymous relationships. In this paper, we propose a synonymous variational inference (SVI) method based on this synonymity viewpoint to re-analyze the perceptual image compression problem. It takes perceptual similarity as a typical synonymous criterion to build an ideal synonymous set (Synset), and approximate the posterior of its latent synonymous representation with a parametric density by minimizing a partial semantic KL divergence. This analysis theoretically proves that the optimization direction of perception image compression follows a triple tradeoff that can cover the existing rate-distortion-perception schemes. Additionally, we introduce synonymous image compression (SIC), a new image compression scheme that corresponds to the analytical process of SVI, and implement a progressive SIC codec to fully leverage the model's capabilities. Experimental results demonstrate comparable rate-distortion-perception performance using a single progressive SIC codec, thus verifying the effectiveness of our proposed analysis method.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 27 of its 52 references that are in the map. First posted on arXiv 28 May 2025.

Builds on

27 of 52 references.

References · 27

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 2018Variational image compression with a scale hyperpriorcited by 935
  5. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 811
  6. 2016End-to-end Optimized Image Compressioncited by 696
  7. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
  8. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 590
  9. 2017The Perception-Distortion Tradeoffcited by 587
  10. 2020High-Fidelity Generative Image Compressioncited by 338
  11. 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 301
  12. 2018Generative Adversarial Networks for Extreme Learned Image Compressioncited by 219
15 more
  1. 2019Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoffcited by 211
  2. 2020Nonlinear Transform Codingcited by 136
  3. 2023Towards image compression with perfect realism at ultra-low bitratescited by 103
  4. 2023Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Modelscited by 94
  5. 2022Lossy Compression with Gaussian Diffusioncited by 90
  6. 2022Multi-Realism Image Compression with a Conditional Generatorcited by 80
  7. 2023Frequency-Aware Transformer for Learned Image Compressioncited by 58
  8. 2023High-Fidelity Image Compression with Score-based Generative Modelscited by 54
  9. 2017Generative Compressioncited by 40
  10. 2022PO-ELIC: Perception-Oriented Efficient Learned Image Codingcited by 36
  11. 2021A coding theorem for the rate-distortion-perception functioncited by 30
  12. 2024Idempotence and Perceptual Image Compressioncited by 29
  13. 2021On the advantages of stochastic encoderscited by 28
  14. 2021On Perceptual Lossy Compression: The Cost of Perceptual Reconstruction and An Optimal Training Frameworkcited by 27
  15. 2023Conditional Perceptual Quality Preserving Image Compressioncited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2026Dual-Representation Image Compression at Ultra-Low Bitrates via Explicit Semantics and Implicit Texturescited by 5
  2. 2026Differentiable Vector Quantization for Rate-Distortion Optimization of Generative Image Compressioncited by 5
  3. 2025Way to Build Native AI-driven 6G Air Interface: Principles, Roadmap, and Outlookcited by 0
  4. 2026A Synonymous Variational Perspective on the Rate-Distortion-Perception Tradeoffcited by 0