mlsanityData to 14 Oct 2026

Lossy Image Compression with Quantized Hierarchical VAEs

arXiv 2208.13056 ↗First posted 27 Aug 2022eess.IV cs.CV cs.LGcited by 35

Not on the map. Why?

Abstract (arXiv)

Recent research has shown a strong theoretical connection between variational autoencoders (VAEs) and the rate-distortion theory. Motivated by this, we consider the problem of lossy image compression from the perspective of generative modeling. Starting with ResNet VAEs, which are originally designed for data (image) distribution modeling, we redesign their latent variable model using a quantization-aware posterior and prior, enabling easy quantization and entropy coding at test time. Along with improved neural network architecture, we present a powerful and efficient model that outperforms previous methods on natural image lossy compression. Our model compresses images in a coarse-to-fine fashion and supports parallel encoding and decoding, leading to fast execution on GPUs. Code is available at https://github.com/duanzhiihao/lossy-vae.

Key facts. Cited by 35 papers in the map (data through 14 Oct 2026). Builds on 43 of its 58 references that are in the map. First posted on arXiv 27 Aug 2022.

Builds on

43 of 58 references.

References · 43

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2017Neural Discrete Representation Learningcited by 4,712
  5. 2013Auto-Encoding Variational Bayescited by 4,053
  6. 2014Deep Learning Face Attributes in the Wildcited by 3,693
  7. 2022A ConvNet for the 2020scited by 3,579
  8. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  9. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  10. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,204
  11. 2018Variational image compression with a scale hyperpriorcited by 935
  12. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
31 more
  1. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  2. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
  3. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 590
  4. 2016Ladder Variational Autoencoderscited by 318
  5. 2017Lossy Image Compression with Compressive Autoencoderscited by 294
  6. 2020CompressAI: a PyTorch library and evaluation platform for end-to-end compression researchcited by 276
  7. 2020Channel-wise Autoregressive Entropy Models for Learned Image Compressioncited by 264
  8. 2020Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Imagescited by 264
  9. 2021Checkerboard Context Model for Efficient Learned Image Compressioncited by 202
  10. 2022The Devil Is in the Details: Window-based Attention for Image Compressioncited by 147
  11. 2020Nonlinear Transform Codingcited by 136
  12. 2016Discrete Variational Autoencoderscited by 125
  13. 2021Enhanced Invertible Encoding for Learned Image Compressioncited by 105
  14. 2019Non-local Attention Optimized Deep Image Compressioncited by 92
  15. 2019Practical Lossless Compression with Latent Variables using Bits Back Codingcited by 86
  16. 2016End-to-end optimization of nonlinear transform codes for perceptual qualitycited by 73
  17. 2020Universally Quantized Neural Compressioncited by 64
  18. 2021Transformer-based Image Compressioncited by 62
  19. 2020Hierarchical Quantized Autoencoderscited by 57
  20. 2019Bit-Swap: Recursive Bits-Back Coding for Lossless Compression with Hierarchical Latent Variablescited by 52
  21. 2021Consistency Regularization for Variational Auto-Encoderscited by 51
  22. 2020Learning Accurate Entropy Model with Global Reference for Image Compressioncited by 45
  23. 2020Compressing Images by Encoding Their Latent Representations with Relative Entropy Codingcited by 39
  24. 2019HiLLoC: Lossless Image Compression with Hierarchical Latent Variable Modelscited by 37
  25. 2021Algorithms for the Communication of Samplescited by 34
  26. 2021On the Out-of-distribution Generalization of Probabilistic Image Modellingcited by 33
  27. 2018DVAE++: Discrete Variational Autoencoders with Overlapping Transformationscited by 32
  28. 2022Neural Data-Dependent Transform for Learned Image Compressioncited by 30
  29. 2018DVAE#: Discrete Variational Autoencoders with Relaxed Boltzmann Priorscited by 20
  30. 2021Towards Empirical Sandwich Bounds on the Rate-Distortion Functioncited by 14
  31. 2020Relaxed-Responsibility Hierarchical Discrete VAEscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 15; 2025: 11; 2026: 3.

35 papers.

Sort:

By area:

  1. 2023MLIC++: Linear Complexity Multi-Reference Entropy Modeling for Learned Image Compressioncited by 58
  2. 2023QARV: Quantization-Aware ResNet VAE for Lossy Image Compressioncited by 32
  3. 2024MambaVC: Learned Visual Compression with Selective State Spacescited by 28
  4. 2024Diffusion-Driven Semantic Communication for Generative Models with Bandwidth Constraintscited by 19
  5. 2024All-in-One Image Coding for Joint Human-Machine Vision with Multi-Path Aggregationcited by 14
  6. 2023Dynamic Low-Rank Instance Adaptation for Universal Neural Image Compressioncited by 10
  7. 2024Progressive Compression with Universally Quantized Diffusion Modelscited by 10
  8. 2025GIViC: Generative Implicit Video Compressioncited by 10
  9. 2024Generalized Nested Latent Variable Models for Lossy Coding applied to Wind Turbine Scenarioscited by 7
  10. 2024Another Way to the Top: Exploit Contextual Clustering in Learned Image Codingcited by 6
  11. 2024STanH : Parametric Quantization for Variable Rate Learned Image Compressioncited by 4
  12. 2024JPEG AI Image Compression Visual Artifacts: Detection Methods and Datasetcited by 4
23 more
  1. 2025Flexible Mixed Precision Quantization for Learned Image Compressioncited by 4
  2. 2024Exploring adversarial robustness of JPEG AI: methodology, comparison and new methodscited by 3
  3. 2024Learned Image Transmission with Hierarchical Variational Autoencodercited by 2
  4. 2024High-Efficiency Neural Video Compression via Hierarchical Predictive Learningcited by 2
  5. 2025Lightweight Embedded FPGA Deployment of Learned Image Compression with Knowledge Distillation and Hybrid Quantizationcited by 2
  6. 2026CodecSplat: Ultra-Compact Latent Coding for Feed-Forward 3D Gaussian Splattingcited by 2
  7. 2023Generative AI for Rapid Diffusion MRI with Improved Image Quality, Reliability and Generalizabilitycited by 1
  8. 2023CPIPS: Learning to Preserve Perceptual Distances in End-to-End Image Compressioncited by 1
  9. 2023Effortless Cross-Platform Video Codec: A Codebook-Based Methodcited by 1
  10. 2024Learned Image Compression with Text Quality Enhancementcited by 1
  11. 2025Gone With the Bits: Revealing Racial Bias in Low-Rate Neural Compression for Facial Imagescited by 1
  12. 2025Structured Pruning and Quantization for Learned Image Compressioncited by 1
  13. 2025NIC-RobustBench: A Comprehensive Open-Source Toolkit for Neural Image Compression and Robustness Analysiscited by 1
  14. 2025Minimal High-Resolution Patches Are Sufficient for Whole Slide Image Representation via Cascaded Dual-Scale Reconstructioncited by 1
  15. 2024CIC: Circular Image Compressioncited by 0
  16. 2024Beta-Sigma VAE: Separating beta and decoder variance in Gaussian variational autoencodercited by 0
  17. 2024Edge-based Denoising Image Compressioncited by 0
  18. 2025Interleaved Block-based Learned Image Compression with Feature Enhancement and Quantization Error Compensationcited by 0
  19. 2025Support is All You Need for Certified VAE Trainingcited by 0
  20. 2025TerraCodec: Compressing Optical Earth Observation Datacited by 0
  21. 2025Adaptive Dataset Quantization: A New Direction for Dataset Pruningcited by 0
  22. 2026ARCHE: Autoregressive Residual Compression with Hyperprior and Excitationcited by 0
  23. 2026Every Packet Counts: Dispersing Information for Loss-Resilient Learned Image Compressioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Lossy Image Compression with Conditional Diffusion Models
  2. 2026Latent-Compressed Variational Autoencoder for Video Diffusion Models
  3. 2025Hierarchical Vector-Quantized Latents for Perceptual Low-Resolution Video Compression
  4. 2023Extreme Image Compression using Fine-tuned VQGANs
  5. 2023Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models
  6. 2026Is Hierarchical Quantization Essential for Optimal Reconstruction?
  7. 2026Tree-VQ: Progressive Image Compression from Pretrained Vector Quantizers
  8. 2025Generative Latent Coding for Ultra-Low Bitrate Image Compression