# HDCompression: Hybrid-Diffusion Image Compression for Ultra-Low Bitrates

arXiv: https://arxiv.org/abs/2502.07160 · HTML: https://mlsanity.com/paper/2502.07160/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion for compression and wireless, one of 14 directions in Diffusion models.
- Builds on 31 of its 50 references that are in the map.
- First posted on arXiv 11 Feb 2025.

## Placed in

- Diffusion models / [Diffusion for compression and wireless](https://mlsanity.com/diffusion/diffusion-for-compression-and-wireless/)

## Builds on (most cited first, 30 of 31)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Conditional Generative Adversarial Nets](https://mlsanity.com/paper/1411.1784/) (2014)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules](https://mlsanity.com/paper/2001.01568/) (2020)
- [The Perception-Distortion Tradeoff](https://mlsanity.com/paper/1711.06077/) (2017)
- [DiffIR: Efficient Diffusion Model for Image Restoration](https://mlsanity.com/paper/2303.09472/) (2023)
- [MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesis](https://mlsanity.com/paper/2211.09117/) (2022)
- [Lossy Image Compression with Conditional Diffusion Models](https://mlsanity.com/paper/2209.06950/) (2022)
- [Return of Unconditional Generation: A Self-supervised Representation Generation Method](https://mlsanity.com/paper/2312.03701/) (2023)
- [Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models](https://mlsanity.com/paper/2301.11189/) (2023)
- [Image and Video Compression with Neural Networks: A Review](https://mlsanity.com/paper/1904.03567/) (2019)
- [MLIC: Multi-Reference Entropy Model for Learned Image Compression](https://mlsanity.com/paper/2211.07273/) (2022)
- [LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers](https://mlsanity.com/paper/2412.12444/) (2024)
- [Lossy Image Compression with Foundation Diffusion Models](https://mlsanity.com/paper/2404.08580/) (2024)
- [CAS-CNN: A Deep Convolutional Neural Network for Image Compression Artifact Suppression](https://mlsanity.com/paper/1611.07233/) (2016)
- [Extreme Image Compression using Fine-tuned VQGANs](https://mlsanity.com/paper/2307.08265/) (2023)
- [SGLP: A Similarity Guided Fast Layer Partition Pruning for Compressing Large Deep Models](https://mlsanity.com/paper/2410.14720/) (2024)
- [Reverse Engineering of Imperceptible Adversarial Image Perturbations](https://mlsanity.com/paper/2203.14145/) (2022)
- [Pruning then Reweighting: Towards Data-Efficient Training of Diffusion Models](https://mlsanity.com/paper/2409.19128/) (2024)
- [HybridFlow: Infusing Continuity into Masked Codebook for Extreme Low-Bitrate Image Compression](https://mlsanity.com/paper/2404.13372/) (2024)
- [Less is More: Data Pruning for Faster Adversarial Training](https://mlsanity.com/paper/2302.12366/) (2023)
- [Learned Lossless Image Compression with a HyperPrior and Discretized Gaussian Mixture Likelihoods](https://mlsanity.com/paper/2002.01657/) (2020)

## Built on by (most cited first, 5 of 5)

- [Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodality](https://mlsanity.com/paper/2505.18227/) (2025)
- [TVC: Tokenized Video Compression with Ultra-Low Bit Rate](https://mlsanity.com/paper/2504.16953/) (2025)
- [Single-step Diffusion for Image Compression at Ultra-Low Bitrates](https://mlsanity.com/paper/2506.16572/) (2025)
- [PECKER: A Precisely Efficient Critical Knowledge Erasure Recipe For Machine Unlearning in Diffusion Models](https://mlsanity.com/paper/2604.05634/) (2026)
- [From Pixels to Nucleotides: End-to-End Token-Based Video Compression for DNA Storage](https://mlsanity.com/paper/2604.13667/) (2026)

## Abstract (arXiv)

Image compression under ultra-low bitrates remains challenging for both conventional learned image compression (LIC) and generative vector-quantized (VQ) modeling. Conventional LIC suffers from severe artifacts due to heavy quantization, while generative VQ modeling gives poor fidelity due to the mismatch between learned generative priors and specific inputs. In this work, we propose Hybrid-Diffusion Image Compression (HDCompression), a dual-stream framework that utilizes both generative VQ-modeling and diffusion models, as well as conventional LIC, to achieve both high fidelity and high perceptual quality. Different from previous hybrid methods that directly use pre-trained LIC models to generate low-quality fidelity-preserving information from heavily quantized latent, we use diffusion models to extract high-quality complementary fidelity information from the ground-truth input, which can enhance the system performance in several aspects: improving index map prediction, enhancing the fidelity-preserving output of the LIC stream, and refining conditioned image reconstruction with VQ-latent correction. In addition, our diffusion model is based on a dense representative vector (DRV), which is lightweight with very simple sampling schedulers. Extensive experiments demonstrate that our HDCompression outperforms the previous conventional LIC, generative VQ-modeling, and hybrid frameworks in both quantitative metrics and qualitative visualization, providing balanced robust compression performance at ultra-low bitrates.