OSCAR: One-Step Diffusion Codec Across Multiple Bit-rates

2025-05-22

AI Summary

Existing diffusion-based image compression methods suffer from two key limitations: high computational cost due to iterative sampling and the need for separate model training per bit-rate, resulting in prohibitive training and storage overhead. This paper proposes OSCAR—the first multi-bit-rate diffusion image compression framework enabling single-step decoding. Its core innovation lies in modeling compressed latent variables as intermediate noise states along a diffusion trajectory and explicitly mapping bit-rate to pseudo-time steps, thereby enabling arbitrary-bit-rate reconstruction via a single model and a single forward pass. OSCAR integrates latent-space compression, structure-aware single-step denoising, conditional time-step embedding, and diffusion prior modeling. Experiments demonstrate that OSCAR consistently outperforms state-of-the-art diffusion-based compression methods in PSNR, LPIPS, and perceptual quality, while accelerating inference by over an order of magnitude and substantially reducing both computational and storage costs.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Pretrained latent diffusion models have shown strong potential for lossy image compression, owing to their powerful generative priors. Most existing diffusion-based methods reconstruct images by iteratively denoising from random noise, guided by compressed latent representations. While these approaches have achieved high reconstruction quality, their multi-step sampling process incurs substantial computational overhead. Moreover, they typically require training separate models for different compression bit-rates, leading to significant training and storage costs. To address these challenges, we propose a one-step diffusion codec across multiple bit-rates. termed OSCAR. Specifically, our method views compressed latents as noisy variants of the original latents, where the level of distortion depends on the bit-rate. This perspective allows them to be modeled as intermediate states along a diffusion trajectory. By establishing a mapping from the compression bit-rate to a pseudo diffusion timestep, we condition a single generative model to support reconstructions at multiple bit-rates. Meanwhile, we argue that the compressed latents retain rich structural information, thereby making one-step denoising feasible. Thus, OSCAR replaces iterative sampling with a single denoising pass, significantly improving inference efficiency. Extensive experiments demonstrate that OSCAR achieves superior performance in both quantitative and visual quality metrics. The code and models will be released at https://github.com/jp-guo/OSCAR.

Problem

• Reduces computational overhead in diffusion-based image compression
• Eliminates need for separate models per bit-rate
• Enables high-quality one-step denoising for efficient reconstruction

Innovation

• One-step denoising for efficient image reconstruction
• Single generative model supports multiple bit-rates
• Compressed latents modeled as diffusion trajectory states