Self-Supervised Image Super-Resolution Quality Assessment based on Content-Free Multi-Model Oriented Representation Learning

2026-02-11

AI Summary

This work addresses the challenge of no-reference image quality assessment for super-resolved images in real-world scenarios, where complex degradation patterns and the absence of ground-truth references or labeled data hinder reliable evaluation. To this end, we propose a self-supervised, no-reference quality assessment method that innovatively treats the super-resolution model itself as a source of degradation, establishing a content-agnostic, multi-model-driven contrastive learning framework. By integrating self-supervised pretraining, tailored preprocessing, and auxiliary tasks, our approach effectively mitigates degradation discrepancies arising from varying scale factors. We further introduce SRMORSS, the first large-scale unlabeled dataset of real-world super-resolved images, to support this paradigm. Extensive experiments demonstrate that our method significantly outperforms existing no-reference metrics across multiple real-world benchmarks, exhibiting strong generalization and robustness under data-scarce conditions.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Unsupervised & Self-Supervised Learning
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Super-resolution (SR) applied to real-world low-resolution (LR) images often results in complex, irregular degradations that stem from the inherent complexity of natural scene acquisition. In contrast to SR artifacts arising from synthetic LR images created under well-defined scenarios, those distortions are highly unpredictable and vary significantly across different real-life contexts. Consequently, assessing the quality of SR images (SR-IQA) obtained from realistic LR, remains a challenging and underexplored problem. In this work, we introduce a no-reference SR-IQA approach tailored for such highly ill-posed realistic settings. The proposed method enables domain-adaptive IQA for real-world SR applications, particularly in data-scarce domains. We hypothesize that degradations in super-resolved images are strongly dependent on the underlying SR algorithms, rather than being solely determined by image content. To this end, we introduce a self-supervised learning (SSL) strategy that first pretrains multiple SR model oriented representations in a pretext stage. Our contrastive learning framework forms positive pairs from images produced by the same SR model and negative pairs from those generated by different methods, independent of image content. The proposed approach S3 RIQA, further incorporates targeted preprocessing to extract complementary quality information and an auxiliary task to better handle the various degradation profiles associated with different SR scaling factors. To this end, we constructed a new dataset, SRMORSS, to support unsupervised pretext training; it includes a wide range of SR algorithms applied to numerous real LR images, which addresses a gap in existing datasets. Experiments on real SR-IQA benchmarks demonstrate that S3 RIQA consistently outperforms most state-of-the-art relevant metrics.

Problem

• Super-resolution
• Image Quality Assessment
• Real-world Degradation
• No-reference
• Self-supervised Learning

Innovation

• self-supervised learning
• super-resolution IQA
• content-free representation
• contrastive learning
• real-world degradation