Studying Aesthetics in Photographic Images Using a Computational Approach

Abstract

Aesthetics, in the world of art and photography, refers to the principles of the nature and appreciation of beauty. Judging beauty and other aesthetic qualities of photographs is a highly subjective task. Hence, there is no unanimously agreed standard for measuring aesthetic value. In spite of the lack of firm rules, certain features in photographic images are believed, by many, to please humans more than certain others. In this paper, we treat the challenge of automatically inferring aesthetic quality of pictures using their visual content as a machine learning problem, with a peer-rated online photo sharing Website as data source. We extract certain visual features based on the intuition that they can discriminate between aesthetically pleasing and displeasing images. Automated classifiers are built using support vector machines and classification trees. Linear regression on polynomial terms of the features is also applied to infer numerical aesthetics ratings. The work attempts to explore the relationship between emotions which pictures arouse in people, and their low-level content. Potential applications include content-based image retrieval and digital photography.

Intelligent Photographing Interface with On-Device Aesthetic Quality Assessment

“…Average Saturation (AS) is considered as another indispensable statistic in computational aesthetics for high-quality photos [2]. It calculates the average value of saturation channel in HSV color space for an image (f 16 ).…”

“…Tong et al [1] attempted to classify photographs into those taken by professionals or home users using low-level features derived from computer vision techniques. Datta et al [2] also employed a set of lowlevel features then followed by a classifier to achieve photo quality assessment. Ke et al [3] designed more semantic features based on the perceptual factors that present the difference between high and low quality photos to increase the performance.…”

Analyzing Emotional Semantics of Abstract Art Using Low-Level Image Features

“…Classification: Support Vector Machine (SVM) [3] is chosen to build the mapping as it is the state-of-art machine learning method and has been used for classification in recent emotion-related studies [19,4]. In our paper, we use the SVM package LIBSVM [2] with default parameters in order to ensure reproducibility of the experimental results.…”

“…[4,9,16,18,19]). Most of the works developed features that are specific to the domains related to art and color theories, which lacks generality and makes it difficult for researchers who are unfamiliar with computer vision algorithms to perform image analysis on their own data [15].…”

Detecting and Removing Visual Distractors for Video Aesthetic Enhancement

“…The users of capture devices also want to improve their results but often do not know how. Thus, researchers in the multimedia community have proposed intelligent methods to evaluate the visual quality of variant media, such as images [3][4] [5], videos [6], paintings [7] and webpages [8] [9]. They both utilize state-of-the-art techniques, such as deep learning [4], to rate aesthetic quality and investigate how the intrinsic features contribute to the final perceptual quality [10] [11].…”