How do Convolutional Neural Networks Learn Design?

被引：0
|
作者：
Jolly, Shailza [1 ]
Iwana, Brian Kenji [2 ]
Kuroki, Ryohei [2 ]
Uchida, Seiichi [2 ]
机构：
[1] Univ Kaiserslautern, Kaiserslautern, Germany
[2] Kyushu Univ, Dept Adv Informat Technol, Fukuoka, Fukuoka, Japan
来源：
2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR) | 2018年
关键词：
CLASSIFICATION;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this paper, we aim to understand the design principles in book cover images which are carefully crafted by experts. Book covers are designed in a unique way, specific to genres which convey important information to their readers. By using Convolutional Neural Networks (CNN) to predict book genres from cover images, visual cues which distinguish genres can be highlighted and analyzed. In order to understand these visual clues contributing towards the decision of a genre, we present the application of Layer-wise Relevance Propagation (LRP) on the book cover image classification results. We use LRP to explain the pixel-wise contributions of book cover design and highlight the design elements contributing towards particular genres. In addition, with the use of state-of-the-art object and text detection methods, insights about genre-specific book cover designs are discovered.
引用
收藏
页码：1085 / 1090
页数：6
相关论文
共 50 条
  • [1] Do Convolutional Neural Networks Learn Class Hierarchy?
    Alsallakh, Bilal
    Jourabloo, Amin
    Ye, Mao
    Liu, Xiaoming
    Ren, Liu
    IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, 2018, 24 (01) : 152 - 162
  • [2] Convolutional neural networks are not invariant to translation, but they can learn to be
    Biscione, Valerio
    Bowers, Jeffrey S.
    Journal of Machine Learning Research, 2021, 22
  • [3] Learn class hierarchy using convolutional neural networks
    La Grassa, Riccardo
    Gallo, Ignazio
    Landro, Nicola
    APPLIED INTELLIGENCE, 2021, 51 (10) : 6622 - 6632
  • [4] Convolutional Neural Networks Are Not Invariant to Translation, but They Can Learn to Be
    Biscione, Valerio
    Bowers, Jeffrey S.
    JOURNAL OF MACHINE LEARNING RESEARCH, 2021, 22
  • [5] Learn class hierarchy using convolutional neural networks
    Riccardo La Grassa
    Ignazio Gallo
    Nicola Landro
    Applied Intelligence, 2021, 51 : 6622 - 6632
  • [6] HOW NEURAL NETWORKS LEARN FROM EXPERIENCE
    HINTON, GE
    SCIENTIFIC AMERICAN, 1992, 267 (03) : 145 - 151
  • [7] Neural networks and how machines learn meaning
    Niculaescu, Oana
    XRDS: Crossroads, 2019, 25 (03): : 64 - 66
  • [8] Convolutional neural networks rarely learn shape for semantic segmentation
    Zhang, Yixin
    Mazurowski, Maciej A.
    PATTERN RECOGNITION, 2024, 146
  • [9] Testing the Ability of Convolutional Neural Networks to Learn Radiomic Features
    Klyuzhin, Ivan S.
    Xu, Yixi
    Ortiz, Anthony
    Ferres, Juan Lavista
    Hamarneh, Ghassan
    Rahmim, Arman
    COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2022, 219
  • [10] Recurrent Convolutional Neural Networks Learn Succinct Learning Algorithms
    Goel, Surbhi
    Kakade, Sham
    Kalai, Adam Tauman
    Zhang, Cyril
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
12345