Automatic aesthetic quality assessment of photographic images using deep convolutional neural network

Abstract

Aesthetics can be defined as the study of various emotions related to the sense of beauty. Nowadays there is a fast growth in the use of digital images which is used for representing information. With the increase in the number of photos, it has become complicated to evaluate its quality. Image aesthetic evaluation methods depends on many aesthetic features. People give more attention to the photos which are visually satisfying. Evaluation of aesthetic quality is a challenging task which require knowledge about aesthetic features. Some low level features like contrast, sharpness, colorfulness etc are used to calculate aesthetic score of an image. This paper present a system for automatic aesthetic quality assessment of photographic images using deep convolutional neural network.