Pixel Perfect: Relational Image Quality Assessment with Spatially-Aware Distortions
For researchers and practitioners in image processing, this method offers an interpretable and annotation-free alternative to traditional IQA, enabling targeted optimization of algorithms.
The paper proposes a relational image quality assessment method that predicts distortion type, intensity, and direction relative to a reference, and outputs a relational quality score without requiring human-labeled data. The approach uses self-supervised synthetic distortions and contrastive learning to achieve interpretable, localized feedback.
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.