Presentation 2003/11/13
Evaluation of the Gender Presumption Capability Performance between Facial Parts
Takuya KAWANO, Kunihito KATO, Kazuhiko YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We use the edge feature and the four directional feature fields on many facial images and then analyzed the feature space. The main purpose of our work is to evaluate which of the subject's facial parts are the most effective at making the difference between males and females. Evaluation of the feature space uses the within-class variance between-class variance ratio and the cluster discriminant analysis. By experimenting on many subjects, we conclude that face, eyes and jaw make the best feature spaces in order to distinguish between male and female subjects.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Gender Differentiation / Facial Parts / Within-class Variance Between-class Variance Ratio / Cluster Discriminant Analysis
Paper # PRMU2003-141,HIP2003-47
Date of Issue

Conference Information
Committee PRMU
Conference Date 2003/11/13(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) Evaluation of the Gender Presumption Capability Performance between Facial Parts
Sub Title (in English)
Keyword(1) Gender Differentiation
Keyword(2) Facial Parts
Keyword(3) Within-class Variance Between-class Variance Ratio
Keyword(4) Cluster Discriminant Analysis
1st Author's Name Takuya KAWANO
1st Author's Affiliation Faculty of Engineering, Gifti University()
2nd Author's Name Kunihito KATO
2nd Author's Affiliation Faculty of Engineering, Gifti University
3rd Author's Name Kazuhiko YAMAMOTO
3rd Author's Affiliation Faculty of Engineering, Gifti University
Date 2003/11/13
Paper # PRMU2003-141,HIP2003-47
Volume (vol) vol.103
Number (no) 452
Page pp.pp.-
#Pages 6
Date of Issue