Presentation 1996/1/18
Surface Curvature-Based Face Recognition using Spherical Correlation
Masaki IKEDA, Hiromi T. TANAKA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Surface curvatures such as Gaussian, mean and principal curvatures are intrinsic surface properties and have played important roles in ccurved surface analysis. In this paper, we propose a face recognition method based on surface curvatures. The shape of a human face are analyzed using principal curvatures and their directions and are represented using the extended gaussian image. Then individual face is recognized by comparing with others using Fisher's spherical correlation. the method is implemented for face range images. Results show that shape information from surface curvatures provides vital cues in discriminating and identifying complex surface structures as human faces.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) 3D curvature / principal direction / face recognition / Extented Gaussian Image / spherical correlation
Paper # PRU95-194
Date of Issue

Conference Information
Committee PRU
Conference Date 1996/1/18(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 Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Surface Curvature-Based Face Recognition using Spherical Correlation
Sub Title (in English)
Keyword(1) 3D curvature
Keyword(2) principal direction
Keyword(3) face recognition
Keyword(4) Extented Gaussian Image
Keyword(5) spherical correlation
1st Author's Name Masaki IKEDA
1st Author's Affiliation Computer Science Department,Ritsumeikan University()
2nd Author's Name Hiromi T. TANAKA
2nd Author's Affiliation Computer Science Department,Ritsumeikan University
Date 1996/1/18
Paper # PRU95-194
Volume (vol) vol.95
Number (no) 469
Page pp.pp.-
#Pages 8
Date of Issue