Manga Face Detection on Various Drawing Styles Using Region Proposals-Based CNN
Issued Date
2023-01-01
Resource Type
eISSN
25869027
Scopus ID
2-s2.0-85151521667
Journal Title
Science and Technology Asia
Volume
28
Issue
1
Start Page
120
End Page
135
Rights Holder(s)
SCOPUS
Bibliographic Citation
Science and Technology Asia Vol.28 No.1 (2023) , 120-135
Suggested Citation
Aukkapinyo K. Manga Face Detection on Various Drawing Styles Using Region Proposals-Based CNN. Science and Technology Asia Vol.28 No.1 (2023) , 120-135. 135. Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/81998
Title
Manga Face Detection on Various Drawing Styles Using Region Proposals-Based CNN
Author(s)
Author's Affiliation
Other Contributor(s)
Abstract
Faces of characters in comic books can be used as meta-features for manga analytics. Manga character faces are not easy for a machine to detect when compared to human faces due to the high variation of drawing styles from various distinct authors. There exist several convolutional neural network-based (CNN-based) frameworks that can achieve high accu-racy in an object detection task. However, their drawback is time and resource consuming to perform data modeling due to the nature of deep learning. Thus, this paper is to propose a method to develop a model using Mask R-CNN, which is one of the CNN-based frameworks, with the transfer learning technique in order to reduce training time and resources while main-taining high performance in the manga character face detection task. The proposed method could achieve the average precision of 87% in the manga character face detection tasks on both seen and unseen drawing styles. It significantly outperforms the existing conventional methods. Moreover, pre-trained weights from MS COCO dataset are transferable to manga character face detection tasks. Therefore, a well-performed manga character face detector could be developed using a limited amount of training data and time.