Pixel Perfect: Relational Image Quality Assessment with Spatially-Aware Distortions

TL;DR

This paper introduces a novel relational and spatially-aware image quality assessment method that uses self-supervised training and contrastive learning to produce interpretable distortion maps without human labels.

Contribution

It proposes a new approach that shifts from absolute quality scores to relational, spatially-aware distortion maps and scores, eliminating the need for manual annotations.

Findings

Produces spatially-aware distortion maps identifying type, intensity, and direction.

Uses contrastive learning on ranked image sets for relational quality scoring.

Does not require human-labeled quality scores, enabling more interpretable IQA.

Abstract

Traditional image quality assessment (IQA) methods rely on mean opinion scores (MOS), which are resource-intensive to collect and fail to provide interpretable, localized feedback on specific image distortions. We overcome these limitations by shifting from absolute quality prediction to a relational and directional assessment. Our approach utilizes a self-supervised synthetic distortion engine to generate training data, eliminating the need for manual annotation. A distortion prediction network is trained with an anti-symmetric objective to produce spatially-aware, disentangled maps that identify the type, intensity, and direction of distortions relative to a reference image. Subsequently, a scoring network is trained via contrastive learning on ordinally ranked image sets to predict a relational quality score. Our method provides a more granular and interpretable approach to IQA for the targeted optimization of image processing algorithms without requiring any human-labeled quality scores.