Explaining Automatic Image Assessment

TL;DR

This paper introduces a method to explain automatic image aesthetic assessment models by visualizing dataset trends and categorizing aesthetic features through neural networks, avoiding manual labeling.

Contribution

It proposes a novel approach that visualizes aesthetic features and trends using neural networks trained on different dataset versions, enhancing explainability.

Findings

Effective visualization of dataset trends and aesthetic features.

Models adapted to each modality show distinct aesthetic patterns.

New metrics successfully evaluate model explanations.

Abstract

Previous work in aesthetic categorization and explainability utilizes manual labeling and classification to explain aesthetic scores. These methods require a complex labeling process and are limited in size. Our proposed approach attempts to explain aesthetic assessment models through visualizing dataset trends and automatic categorization of visual aesthetic features through training neural networks on different versions of the same dataset. By evaluating the models adapted to each specific modality using existing and novel metrics, we can capture and visualize aesthetic features and trends.