Next-Frame Decoding for Ultra-Low-Bitrate Image Compression with Video Diffusion Priors

TL;DR

This paper introduces a new ultra-low-bitrate image compression method that uses a video diffusion prior to improve image quality by modeling the transition from a compact anchor frame to the final image.

Contribution

It proposes a novel decoding paradigm leveraging a pretrained video diffusion model to enhance perceptual quality at ultra-low bitrates with a semantic anchor frame.

Findings

Achieves over 50% bitrate savings on CLIC2020 test set.

Improves perceptual quality and realism compared to previous diffusion-based methods.

Decoding speed is increased by up to 5 times.

Abstract

We present a novel paradigm for ultra-low-bitrate image compression (ULB-IC) that exploits the "temporal" evolution in generative image compression. Specifically, we define an explicit intermediate state during decoding: a compact anchor frame, which preserves the scene geometry and semantic layout while discarding high-frequency details. We then reinterpret generative decoding as a virtual temporal transition from this anchor to the final reconstructed image.To model this progression, we leverage a pretrained video diffusion model (VDM) as temporal priors: the anchor frame serves as the initial frame and the original image as the target frame, transforming the decoding process into a next-frame prediction task.In contrast to image diffusion-based ULB-IC models, our decoding proceeds from a visible, semantically faithful anchor, which improves both fidelity and realism for perceptual image compression. Extensive experiments demonstrate that our method achieves superior objective and subjective performance. On the CLIC2020 test set, our method achieves over 50% bitrate savings across LPIPS, DISTS, FID, and KID compared to DiffC, while also delivering a significant decoding speedup of up to $\times$5. Code will be released later.