Presentation 2015-03-20
Discussing of learning method for labeling manga characters
Eiki IMAZU, Motoi IWATA, Koichi KISE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, I propose a new method to train similarity for labeling characters' ID to manga character regions. In manga, manga characters are important elements. If they can be retrieved, it's possible to meet user's demand. Labeling to manga characters is difficult with simple pattern matching because the manga characters have the various directions of the faces and expressions. Conventional method's problem is labeling many manga characters. In addition, human have to choose many kinds of face data. In this paper, the proposed method employs pool-based active learning to solves these problems. Pool-based active learning has an advantage to solve the problem when obtaining samples with labels requires quite higher cost than the cost to obtain samples without labels. I used our method for labeling experiment and show its advantages.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) manga / character retrieval / pool-based active learning
Paper # BioX2014-65,PRMU2014-185
Date of Issue

Conference Information
Committee PRMU
Conference Date 2015/3/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Discussing of learning method for labeling manga characters
Sub Title (in English)
Keyword(1) manga
Keyword(2) character retrieval
Keyword(3) pool-based active learning
1st Author's Name Eiki IMAZU
1st Author's Affiliation Graduate School of Engineering, Osaka Prefecture University()
2nd Author's Name Motoi IWATA
2nd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
3rd Author's Name Koichi KISE
3rd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
Date 2015-03-20
Paper # BioX2014-65,PRMU2014-185
Volume (vol) vol.114
Number (no) 521
Page pp.pp.-
#Pages 6
Date of Issue