Content-Diverse Comparisons improve IQA

TL;DR

This paper enhances image quality assessment by incorporating content-diverse comparisons and listwise evaluation, leading to improved model training across various benchmarks.

Contribution

It introduces a novel training scheme that relaxes comparison constraints and uses listwise comparisons with regularizers, increasing diversity and holistic evaluation in IQA models.

Findings

Improved correlation with human judgments across benchmarks.

Enhanced model robustness to diverse distortions.

Better score consistency through listwise comparison techniques.

Abstract

Image quality assessment (IQA) forms a natural and often straightforward undertaking for humans, yet effective automation of the task remains highly challenging. Recent metrics from the deep learning community commonly compare image pairs during training to improve upon traditional metrics such as PSNR or SSIM. However, current comparisons ignore the fact that image content affects quality assessment as comparisons only occur between images of similar content. This restricts the diversity and number of image pairs that the model is exposed to during training. In this paper, we strive to enrich these comparisons with content diversity. Firstly, we relax comparison constraints, and compare pairs of images with differing content. This increases the variety of available comparisons. Secondly, we introduce listwise comparisons to provide a holistic view to the model. By including differentiable regularizers, derived from correlation coefficients, models can better adjust predicted scores relative to one another. Evaluation on multiple benchmarks, covering a wide range of distortions and image content, shows the effectiveness of our learning scheme for training image quality assessment models.