Steering One-Step Diffusion Model with Fidelity-Rich Decoder for Fast Image Compression
This work addresses the critical drawbacks of slow decoding and low fidelity in diffusion-based image compression, offering a faster and more faithful solution for applications requiring efficient image compression.
The paper tackled the problems of excessive decoding latency and poor fidelity in diffusion-based image compression by proposing SODEC, a single-step model that uses a fidelity-rich decoder, achieving a more than 20x speed improvement in decoding speed while delivering superior rate-distortion-perception performance.
Diffusion-based image compression has demonstrated impressive perceptual performance. However, it suffers from two critical drawbacks: (1) excessive decoding latency due to multi-step sampling, and (2) poor fidelity resulting from over-reliance on generative priors. To address these issues, we propose SODEC, a novel single-step diffusion image compression model. We argue that in image compression, a sufficiently informative latent renders multi-step refinement unnecessary. Based on this insight, we leverage a pre-trained VAE-based model to produce latents with rich information, and replace the iterative denoising process with a single-step decoding. Meanwhile, to improve fidelity, we introduce the fidelity guidance module, encouraging output that is faithful to the original image. Furthermore, we design the rate annealing training strategy to enable effective training under extremely low bitrates. Extensive experiments show that SODEC significantly outperforms existing methods, achieving superior rate-distortion-perception performance. Moreover, compared to previous diffusion-based compression models, SODEC improves decoding speed by more than 20$\times$. Code is released at: https://github.com/zhengchen1999/SODEC.