FaithDiff: Unleashing Diffusion Priors for Faithful Image Super-resolution

TL;DR

FaithDiff leverages latent diffusion models with a novel alignment and joint fine-tuning approach to achieve highly faithful and structurally consistent image super-resolution, outperforming existing methods.

Contribution

The paper introduces FaithDiff, a method that fully exploits diffusion priors with joint fine-tuning and alignment for faithful image super-resolution.

Findings

FaithDiff surpasses state-of-the-art SR methods in fidelity and quality.

Joint fine-tuning of encoder and diffusion model enhances structural consistency.

Alignment module effectively bridges degraded inputs and diffusion features.

Abstract

Faithful image super-resolution (SR) not only needs to recover images that appear realistic, similar to image generation tasks, but also requires that the restored images maintain fidelity and structural consistency with the input. To this end, we propose a simple and effective method, named FaithDiff, to fully harness the impressive power of latent diffusion models (LDMs) for faithful image SR. In contrast to existing diffusion-based SR methods that freeze the diffusion model pre-trained on high-quality images, we propose to unleash the diffusion prior to identify useful information and recover faithful structures. As there exists a significant gap between the features of degraded inputs and the noisy latent from the diffusion model, we then develop an effective alignment module to explore useful features from degraded inputs to align well with the diffusion process. Considering the indispensable roles and interplay of the encoder and diffusion model in LDMs, we jointly fine-tune them in a unified optimization framework, facilitating the encoder to extract useful features that coincide with diffusion process. Extensive experimental results demonstrate that FaithDiff outperforms state-of-the-art methods, providing high-quality and faithful SR results.