Presentation 1999/6/17
Updating Face Dictionary against Temporal Change
Osamu Yamaguchi, Kazuhiro Fukui,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a method of updating face dictionary in a face recognition system using an image sequence. Update of face dictionary is necessary for personal identification to alleviiate the influence of temporal facial clange, since the performance of recognition gradually declines for the reasons such as aging or changes in hair style. We have first collected numerous data during several months and evaluated the infltuence of temporal facial change on the system. In order to keep the high recognition performance, we have examinned methods of updating the dictionary, including addition of new face data, deletion of the old ones, and application of reinforced learning. The effectiveness of these dictionary updates is shown through the expental results.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Face Recognition / Image Sequence / Temporal Change / Face Dictionary / Update / Mutural Subspace Method / Reinforced Learning
Paper # PRMU99-25
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/6/17(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) Updating Face Dictionary against Temporal Change
Sub Title (in English)
Keyword(1) Face Recognition
Keyword(2) Image Sequence
Keyword(3) Temporal Change
Keyword(4) Face Dictionary
Keyword(5) Update
Keyword(6) Mutural Subspace Method
Keyword(7) Reinforced Learning
1st Author's Name Osamu Yamaguchi
1st Author's Affiliation TOSHIBA Kansai Research Center()
2nd Author's Name Kazuhiro Fukui
2nd Author's Affiliation TOSHIBA Kansai Research Center
Date 1999/6/17
Paper # PRMU99-25
Volume (vol) vol.99
Number (no) 118
Page pp.pp.-
#Pages 8
Date of Issue