IRBridge: Solving Image Restoration Bridge with Pre-trained Generative Diffusion Models

TL;DR

IRBridge leverages pretrained generative diffusion models to enhance image restoration, reducing the need for task-specific training and improving robustness across multiple restoration tasks.

Contribution

The paper introduces IRBridge, a novel framework that integrates pretrained generative priors into image restoration bridges, addressing distribution mismatch and enabling flexible, efficient restoration.

Findings

Improved robustness and generalization in six image restoration tasks.

Efficient utilization of pretrained generative models without retraining for each task.

Enhanced performance over traditional bridge models in image restoration.

Abstract

Bridge models in image restoration construct a diffusion process from degraded to clear images. However, existing methods typically require training a bridge model from scratch for each specific type of degradation, resulting in high computational costs and limited performance. This work aims to efficiently leverage pretrained generative priors within existing image restoration bridges to eliminate this requirement. The main challenge is that standard generative models are typically designed for a diffusion process that starts from pure noise, while restoration tasks begin with a low-quality image, resulting in a mismatch in the state distributions between the two processes. To address this challenge, we propose a transition equation that bridges two diffusion processes with the same endpoint distribution. Based on this, we introduce the IRBridge framework, which enables the direct utilization of generative models within image restoration bridges, offering a more flexible and adaptable approach to image restoration. Extensive experiments on six image restoration tasks demonstrate that IRBridge efficiently integrates generative priors, resulting in improved robustness and generalization performance. Code will be available at GitHub.