Exploring Kolmogorov-Arnold networks for realistic image sharpness assessment

TL;DR

This paper introduces TaylorKAN, a novel Kolmogorov-Arnold network variant, for realistic image sharpness assessment, demonstrating its effectiveness across multiple datasets and outperforming support vector regression.

Contribution

It is the first study to apply and analyze KANs for image quality assessment, proposing a Taylor series-based KAN and exploring its performance with various features.

Findings

TaylorKAN outperforms support vector regression in image sharpness prediction.

KANs are generally competitive or superior to existing methods.

Mid-level features enhance the accuracy of KAN-based assessments.

Abstract

Score prediction is crucial in evaluating realistic image sharpness based on collected informative features. Recently, Kolmogorov-Arnold networks (KANs) have been developed and witnessed remarkable success in data fitting. This study introduces the Taylor series-based KAN (TaylorKAN). Then, different KANs are explored in four realistic image databases (BID2011, CID2013, CLIVE, and KonIQ-10k) to predict the scores by using 15 mid-level features and 2048 high-level features. Compared to support vector regression, results show that KANs are generally competitive or superior, and TaylorKAN is the best one when mid-level features are used. This is the first study to investigate KANs on image quality assessment that sheds some light on how to select and further improve KANs in related tasks.