Despite the tremendous success of diffusion generative models in text-to-image generation, replicating this success in the domain of image compression has proven difficult. In this paper, we demonstrate that diffusion can significantly improve perceptual quality at a given bit-rate, outperforming state-of-the-art approaches PO-ELIC and HiFiC as measured by FID score. This is achieved using a simple but theoretically motivated two-stage approach combining an autoencoder targeting MSE followed by a further score-based decoder. However, as we will show, implementation details matter and the optimal design decisions can differ greatly from typical text-to-image models.
@article{arxiv.2305.18231,
title = {High-Fidelity Image Compression with Score-based Generative Models},
author = {Emiel Hoogeboom and Eirikur Agustsson and Fabian Mentzer and Luca Versari and George Toderici and Lucas Theis},
journal= {arXiv preprint arXiv:2305.18231},
year = {2024}
}