Lossy Image Compression with Conditional Diffusion Models

Ruihan Yang, Stephan Mandt

Keywords: neural image compression, denoising diffusion models

Abstract: Denoising diffusion models have recently marked a milestone in high-quality image generation. One may thus wonder if they are suitable for neural image compression. This paper outlines an end-to-end optimized image compression framework based on a conditional diffusion model, drawing on the transform-coding paradigm. Besides the latent variables inherent to the diffusion process, this paper introduces an additional discrete "content" latent variable to condition the denoising process on. This variable is equipped with a hierarchical prior for entropy coding. The remaining "texture" latent variables characterizing the diffusion process are synthesized (either stochastically or deterministically) at decoding time. We furthermore show that the performance can be tuned toward perceptual metrics of interest. Our extensive experiments involving five datasets and 16 image perceptual quality assessment metrics show that our approach not only compares favorably in terms of rate and perceptual distortion tradeoffs but also shows robust performance under all metrics while other baselines show less consistent behavior.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Generative models

TL;DR: We show that hybridizing compressive VAEs with denoising diffusion models leads to strong performance in perceptual image compression..

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2209.06950/code)