Enhanced Control for Diffusion Bridge in Image Restoration

TL;DR

This paper introduces the ECDB model that enhances control in diffusion bridge-based image restoration, achieving state-of-the-art results across multiple tasks by incorporating conditional control and a novel fusion schedule.

Contribution

The paper proposes the ECDB model with improved conditional control and a Conditional Fusion Schedule for diffusion-based image restoration.

Findings

ECDB achieves state-of-the-art results in deraining, inpainting, and super-resolution.

The Conditional Fusion Schedule improves handling of conditional features.

ECDB outperforms existing diffusion bridge models in image restoration tasks.

Abstract

Image restoration refers to the process of restoring a damaged low-quality image back to its corresponding high-quality image. Typically, we use convolutional neural networks to directly learn the mapping from low-quality images to high-quality images achieving image restoration. Recently, a special type of diffusion bridge model has achieved more advanced results in image restoration. It can transform the direct mapping from low-quality to high-quality images into a diffusion process, restoring low-quality images through a reverse process. However, the current diffusion bridge restoration models do not emphasize the idea of conditional control, which may affect performance. This paper introduces the ECDB model enhancing the control of the diffusion bridge with low-quality images as conditions. Moreover, in response to the characteristic of diffusion models having low denoising level at larger values of \(\bm t \), we also propose a Conditional Fusion Schedule, which more effectively handles the conditional feature information of various modules. Experimental results prove that the ECDB model has achieved state-of-the-art results in many image restoration tasks, including deraining, inpainting and super-resolution. Code is avaliable at https://github.com/Hammour-steak/ECDB.