DP$^2$O-SR: Direct Perceptual Preference Optimization for Real-World Image Super-Resolution

TL;DR

This paper introduces DP$^2$O-SR, a novel framework that optimizes perceptual quality in real-world image super-resolution by leveraging the stochasticity of diffusion models and a hybrid reward system, without requiring human annotations.

Contribution

It proposes a new perceptual preference optimization method that uses multiple preference pairs and hierarchical weighting to improve super-resolution quality without costly human labels.

Findings

Significant improvement in perceptual quality on real-world benchmarks.

Effective utilization of stochastic diversity in diffusion models.

Adaptive preference weighting enhances training stability.

Abstract

Benefiting from pre-trained text-to-image (T2I) diffusion models, real-world image super-resolution (Real-ISR) methods can synthesize rich and realistic details. However, due to the inherent stochasticity of T2I models, different noise inputs often lead to outputs with varying perceptual quality. Although this randomness is sometimes seen as a limitation, it also introduces a wider perceptual quality range, which can be exploited to improve Real-ISR performance. To this end, we introduce Direct Perceptual Preference Optimization for Real-ISR (DP$^2$O-SR), a framework that aligns generative models with perceptual preferences without requiring costly human annotations. We construct a hybrid reward signal by combining full-reference and no-reference image quality assessment (IQA) models trained on large-scale human preference datasets. This reward encourages both structural fidelity and natural appearance. To better utilize perceptual diversity, we move beyond the standard best-vs-worst selection and construct multiple preference pairs from outputs of the same model. Our analysis reveals that the optimal selection ratio depends on model capacity: smaller models benefit from broader coverage, while larger models respond better to stronger contrast in supervision. Furthermore, we propose hierarchical preference optimization, which adaptively weights training pairs based on intra-group reward gaps and inter-group diversity, enabling more efficient and stable learning. Extensive experiments across both diffusion- and flow-based T2I backbones demonstrate that DP$^2$O-SR significantly improves perceptual quality and generalizes well to real-world benchmarks.