Predicting Aesthetic Score Distribution through Cumulative Jensen-Shannon Divergence

TL;DR

This paper introduces a novel deep learning approach, CJS-CNN, to predict aesthetic score distributions from images, capturing human perception more accurately than scalar ratings by using a cumulative Jensen-Shannon divergence method.

Contribution

The paper proposes a new CNN model based on cumulative Jensen-Shannon divergence for predicting aesthetic score distributions, incorporating a kurtosis-based reliability-sensitive learning method.

Findings

CJS-CNN outperforms existing methods on large-scale aesthetic datasets.

The approach effectively models the variability in human aesthetic ratings.

Reliability-sensitive learning improves prediction accuracy without needing full human rating data.

Abstract

Aesthetic quality prediction is a challenging task in the computer vision community because of the complex interplay with semantic contents and photographic technologies. Recent studies on the powerful deep learning based aesthetic quality assessment usually use a binary high-low label or a numerical score to represent the aesthetic quality. However the scalar representation cannot describe well the underlying varieties of the human perception of aesthetics. In this work, we propose to predict the aesthetic score distribution (i.e., a score distribution vector of the ordinal basic human ratings) using Deep Convolutional Neural Network (DCNN). Conventional DCNNs which aim to minimize the difference between the predicted scalar numbers or vectors and the ground truth cannot be directly used for the ordinal basic rating distribution. Thus, a novel CNN based on the Cumulative distribution with Jensen-Shannon divergence (CJS-CNN) is presented to predict the aesthetic score distribution of human ratings, with a new reliability-sensitive learning method based on the kurtosis of the score distribution, which eliminates the requirement of the original full data of human ratings (without normalization). Experimental results on large scale aesthetic dataset demonstrate the effectiveness of our introduced CJS-CNN in this task.