Brain-Inspired Deep Networks for Image Aesthetics Assessment

TL;DR

This paper introduces Brain-Inspired Deep Networks (BDN) for image aesthetics assessment, leveraging neuroaesthetic principles to improve prediction accuracy and handle subjective rating distributions.

Contribution

The paper presents a novel biologically inspired deep network architecture that learns attributes and predicts aesthetic ratings, including rating distributions, with improved performance.

Findings

Significant performance gains on the AVA dataset.

Effective data augmentation using label-preserving transformations.

First study of label-preserving transformations in aesthetics assessment.

Abstract

Image aesthetics assessment has been challenging due to its subjective nature. Inspired by the scientific advances in the human visual perception and neuroaesthetics, we design Brain-Inspired Deep Networks (BDN) for this task. BDN first learns attributes through the parallel supervised pathways, on a variety of selected feature dimensions. A high-level synthesis network is trained to associate and transform those attributes into the overall aesthetics rating. We then extend BDN to predicting the distribution of human ratings, since aesthetics ratings are often subjective. Another highlight is our first-of-its-kind study of label-preserving transformations in the context of aesthetics assessment, which leads to an effective data augmentation approach. Experimental results on the AVA dataset show that our biological inspired and task-specific BDN model gains significantly performance improvement, compared to other state-of-the-art models with the same or higher parameter capacity.