DeltaDiff: Reality-Driven Diffusion with AnchorResiduals for Faithful SR

TL;DR

DeltaDiff introduces a deterministic diffusion framework for super-resolution that reduces noise and improves fidelity, outperforming existing models in generating high-quality images.

Contribution

The paper presents a novel low-rank constrained diffusion method that enhances fidelity in super-resolution by establishing a deterministic mapping between LR and HR images.

Findings

25% reduction in diffusion entropy in residual space

Outperforms state-of-the-art super-resolution models

Generates images with better fidelity

Abstract

Recently, the transfer application of diffusion models in super-resolu-tion tasks has faced the problem ofdecreased fidelity. Due to the inherent randomsampling characteristics ofdiffusion models, direct application in super-resolu-tion tasks can result in generated details deviating from the true distribution ofhigh-resolution images. To address this, we propose DeltaDiff, a novel frame.work that constrains the difusion process, its essence is to establish a determin-istic mapping path between HR and LR, rather than the random noise disturbanceprocess oftraditional difusion models. Theoretical analysis demonstrates a 25%reduction in diffusion entropy in the residual space compared to pixel-space diffiusion, effectively suppressing irrelevant noise interference. The experimentalresults show that our method surpasses state-of-the-art models and generates re-sults with better fidelity. This work establishes a new low-rank constrained par-adigm for applying diffusion models to image reconstruction tasks, balancingstochastic generation with structural fidelity. Our code and model are publiclyavailable at https://github.com/continueyang/DeltaDiff .