FiDeSR: High-Fidelity and Detail-Preserving One-Step Diffusion Super-Resolution

TL;DR

FiDeSR is a novel diffusion-based super-resolution framework that enhances image details and fidelity through adaptive strategies and residual noise refinement, outperforming existing methods in real-world scenarios.

Contribution

The paper introduces FiDeSR, a one-step diffusion super-resolution method with adaptive detail emphasis and residual noise correction, improving detail preservation and high-fidelity reconstruction.

Findings

Outperforms existing diffusion-based SR methods in real-world scenarios

Produces high perceptual quality and faithful content restoration

Incorporates adaptive enhancement and residual noise refinement

Abstract

Diffusion-based approaches have recently driven remarkable progress in real-world image super-resolution (SR). However, existing methods still struggle to simultaneously preserve fine details and ensure high-fidelity reconstruction, often resulting in suboptimal visual quality. In this paper, we propose FiDeSR, a high-fidelity and detail-preserving one-step diffusion super-resolution framework. During training, we introduce a detail-aware weighting strategy that adaptively emphasizes regions where the model exhibits higher prediction errors. During inference, low- and high-frequency adaptive enhancers further refine the reconstruction without requiring model retraining, enabling flexible enhancement control. To further improve the reconstruction accuracy, FiDeSR incorporates a residual-in-residual noise refinement, which corrects prediction errors in the diffusion noise and enhances fine detail recovery. FiDeSR achieves superior real-world SR performance compared to existing diffusion-based methods, producing outputs with both high perceptual quality and faithful content restoration. The source code will be released at: https://github.com/Ar0Kim/FiDeSR.