Presentation 1998/11/13
A face recognition algorithm invariant to object pose and light direction based on geo-photometric constraint
Satoshi Haruyama, Hitoshi Sakano, Naoki Mukawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Face recognition is one of the main issues in the field of pattern recognition. Existence face recognition algorithms do not accurately recognize if its pose and illumination vary. In this paper we propose a novel face recognition algorithm which provides a system that is invariant to the variation of object pose and light direction. We empoly geometric and photometric constraints. to assure robustness against illumination and face pose variation. The method is tested on a computer graphics images of 10 faces, and the recognition rate was 89%.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) face recognition / geotensity consraint structure from motion
Paper # PRMU98-134
Date of Issue

Conference Information
Committee PRMU
Conference Date 1998/11/13(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) A face recognition algorithm invariant to object pose and light direction based on geo-photometric constraint
Sub Title (in English)
Keyword(1) face recognition
Keyword(2) geotensity consraint structure from motion
1st Author's Name Satoshi Haruyama
1st Author's Affiliation Laboratry for Information Technology NTT DATA CORPORATION()
2nd Author's Name Hitoshi Sakano
2nd Author's Affiliation Laboratry for Information Technology NTT DATA CORPORATION
3rd Author's Name Naoki Mukawa
3rd Author's Affiliation Laboratry for Information Technology NTT DATA CORPORATION
Date 1998/11/13
Paper # PRMU98-134
Volume (vol) vol.98
Number (no) 395
Page pp.pp.-
#Pages 8
Date of Issue