BIROn - Birkbeck Institutional Research Online

    Hierarchical aesthetic quality assessment using deep convolutional neural networks

    Kao, Y. and Huang, K. and Maybank, Stephen (2016) Hierarchical aesthetic quality assessment using deep convolutional neural networks. Signal Processing: Image Communication 47 (C), pp. 500-510. ISSN 0923-5965.

    [img]
    Preview
    Text
    15178.pdf - Author's Accepted Manuscript

    Download (1MB) | Preview

    Abstract

    Aesthetic image analysis has attracted much attention in recent years. However, assessing the aesthetic quality and assigning an aesthetic score are challenging problems. In this paper, we propose a novel framework for assessing the aesthetic quality of images. Firstly, we divide the images into three categories: “scene”, “object” and “texture”. Each category has an associated convolutional neural network (CNN) which learns the aesthetic features for the category in question. The object CNN is trained using the whole images and a salient region in each image. The texture CNN is trained using small regions in the original images. Furthermore, an A & C CNN is developed to simultaneously assess the aesthetic quality and identify the category for overall images. For each CNN, classification and regression models are developed separately to predict aesthetic class (high or low) and to assign an aesthetic score. Experimental results on a recently published large-scale dataset show that the proposed method can outperform the state-of-the-art methods for each category.

    Metadata

    Item Type: Article
    Keyword(s) / Subject(s): Aesthetic image analysis, Convolutional neural networks, Scene, Object, Texture
    School: Birkbeck Schools and Departments > School of Business, Economics & Informatics > Computer Science and Information Systems
    Depositing User: Administrator
    Date Deposited: 16 May 2016 12:35
    Last Modified: 26 Jul 2019 17:44
    URI: http://eprints.bbk.ac.uk/id/eprint/15178

    Statistics

    Downloads
    Activity Overview
    154Downloads
    147Hits

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item