Presentation 2002/11/15
Parallel Optimization of Class Configuration and Feature Space for Object Recognition
Mihoko SHIMANO, Kenji NAGAO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a new method to classify objects in images into categories explicitly specified by applications. In many object recognition methods, these categories define the classes for supervised classification themselves. In general, however, separability of such classes isn't guaranteed. A solution to this problem has been found that combines Fisher's separability criterion and information criterion of AIC to optimize the class configuration and the feature space, increasing the class-separability. Effectiveness of the new method will be demonstrated using real images of human faces.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) object recognition / Fisher's discirminant analysis / visual learning / AIC
Paper # HIP2002-35
Date of Issue

Conference Information
Committee HIP
Conference Date 2002/11/15(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 Human Information Processing (HIP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Parallel Optimization of Class Configuration and Feature Space for Object Recognition
Sub Title (in English)
Keyword(1) object recognition
Keyword(2) Fisher's discirminant analysis
Keyword(3) visual learning
Keyword(4) AIC
1st Author's Name Mihoko SHIMANO
1st Author's Affiliation Matsushita Electric Industrial Co., Ltd. Advanced Technology Research Laboratories()
2nd Author's Name Kenji NAGAO
2nd Author's Affiliation Matsushita Electric Industrial Co., Ltd. Advanced Technology Research Laboratories
Date 2002/11/15
Paper # HIP2002-35
Volume (vol) vol.102
Number (no) 473
Page pp.pp.-
#Pages 6
Date of Issue