Regression-free Blind Image Quality Assessment with Content-Distortion Consistency

TL;DR

This paper introduces a regression-free image quality assessment framework that leverages content-distortion consistency and instance retrieval, achieving competitive results without training on subjective scores.

Contribution

It proposes a novel regression-free approach using hierarchical k-NN for content and distortion similarity, avoiding training biases of traditional regression-based IQA models.

Findings

Achieves superior performance on IQA benchmarks without training on subjective scores.

Utilizes content-distortion consistency inspired by human visual perception.

Outperforms state-of-the-art regression-based methods in accuracy.

Abstract

The optimization objective of regression-based blind image quality assessment (IQA) models is to minimize the mean prediction error across the training dataset, which can lead to biased parameter estimation due to potential training data biases. To mitigate this issue, we propose a regression-free framework for image quality evaluation, which is based upon retrieving locally similar instances by incorporating semantic and distortion feature spaces. The approach is motivated by the observation that the human visual system (HVS) exhibits analogous perceptual responses to semantically similar image contents impaired by identical distortions, which we term as content-distortion consistency. The proposed method constructs a hierarchical k-nearest neighbor (k-NN) algorithm for instance retrieval through two classification modules: semantic classification (SC) module and distortion classification (DC) module. Given a test image and an IQA database, the SC module retrieves multiple pristine images semantically similar to the test image. The DC module then retrieves instances based on distortion similarity from the distorted images that correspond to each retrieved pristine image. Finally, quality prediction is obtained by aggregating the subjective scores of the retrieved instances. Without training on subjective quality scores, the proposed regression-free method achieves competitive, even superior performance compared to state-of-the-art regression-based methods on authentic and synthetic distortion IQA benchmarks.