Bridging Fidelity-Reality with Controllable One-Step Diffusion for Image Super-Resolution

TL;DR

This paper introduces CODSR, a controllable one-step diffusion model for image super-resolution that improves fidelity, perceptual quality, and semantic alignment by leveraging uncompressed input information, region-adaptive priors, and text guidance.

Contribution

The paper proposes a novel controllable diffusion framework with LQ-guided modulation, region-adaptive priors, and text guidance to address fidelity and semantic alignment issues in super-resolution.

Findings

Achieves superior perceptual quality over state-of-the-art methods.

Maintains competitive fidelity with efficient one-step inference.

Effectively aligns text prompts with semantic regions in images.

Abstract

Recent diffusion-based one-step methods have shown remarkable progress in the field of image super-resolution, yet they remain constrained by three critical limitations: (1) inferior fidelity performance caused by the information loss from compression encoding of low-quality (LQ) inputs; (2) insufficient region-discriminative activation of generative priors; (3) misalignment between text prompts and their corresponding semantic regions. To address these limitations, we propose CODSR, a controllable one-step diffusion network for image super-resolution. First, we propose an LQ-guided feature modulation module that leverages original uncompressed information from LQ inputs to provide high-fidelity conditioning for the diffusion process. We then develop a region-adaptive generative prior activation method to effectively enhance perceptual richness without sacrificing local structural fidelity. Finally, we employ a text-matching guidance strategy to fully harness the conditioning potential of text prompts. Extensive experiments demonstrate that CODSR achieves superior perceptual quality and competitive fidelity compared with state-of-the-art methods with efficient one-step inference.