Feature visualization in comic artist classification using deep neural networks

Autor: Kim Young-Min
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Information Systems and Management
lcsh:Computer engineering. Computer hardware
Computer Networks and Communications
Computer science
lcsh:TK7885-7895
02 engineering and technology
Input format
Comics
Convolutional neural network
lcsh:QA75.5-76.95
020204 information systems
Deep neural networks
0202 electrical engineering
electronic engineering
information engineering
Artistic styles
lcsh:T58.5-58.64
business.industry
lcsh:Information technology
Pattern recognition
Visualization
ComputingMethodologies_PATTERNRECOGNITION
Hardware and Architecture
Analytics
Comic classification
Feature visualization
020201 artificial intelligence & image processing
Convolutional neural networks
Artificial intelligence
lcsh:Electronic computers. Computer science
F1 score
business
Classifier (UML)
Information Systems
Zdroj: Journal of Big Data, Vol 6, Iss 1, Pp 1-18 (2019)
ISSN: 2196-1115
DOI: 10.1186/s40537-019-0222-3
Popis: Deep neural networks have become a standard framework for image analytics. Besides the traditional applications, such as object classification and detection, the latest studies have started to expand the scope of the applications to include artworks. However, popular art forms, such as comics, have been ignored in this trend. This study investigates visual features for comic classification using deep neural networks. An effective input format for comic classification is first defined, and a convolutional neural network is used to classify comic images into eight different artist categories. Using a publicly available dataset, the trained model obtains a mean F1 score of 84% for the classification. A feature visualization technique is also applied to the trained classifier, to verify the internal visual characteristics that succeed in classification. The experimental result shows that the visualized features are significantly different from those of general object classification. This work represents one of the first attempts to examine the visual characteristics of comics using feature visualization, in terms of comic author classification with deep neural networks.
Databáze: OpenAIRE
Externí odkaz: