Presentation 2006-03-16
An Approach to Adaptive and Trainable Vision
Nobuyuki OTSU,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper (invited talk) presents an outline of our approach and research results towards adaptive and trainable vision systems which the author has been devoted to. First, such an approach is emphasized that is based on the general framework of pattern recognition consisting of two stages of feature extraction; invariant feature extraction as the first geometrical aspect, and discriminant feature extraction as the second statistical aspect. Along the line of the general approach, a scheme of general-purpose adaptive and trainable image recognition comprising Higher-order Local Auto-Correlation (HLAC) feature extraction and multivariate analysis methods so far developed and its several applications are introduced. Finally, some recent researches using extended HLAC (CHLAC) features for several practical applications of motion recognition are presented.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) vision system / pattern recognition / feature extraction / multivariate analysis
Paper # PRMU2005-258
Date of Issue

Conference Information
Committee PRMU
Conference Date 2006/3/9(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) An Approach to Adaptive and Trainable Vision
Sub Title (in English)
Keyword(1) vision system
Keyword(2) pattern recognition
Keyword(3) feature extraction
Keyword(4) multivariate analysis
1st Author's Name Nobuyuki OTSU
1st Author's Affiliation National Institute of Advanced Industrial Science and Technology:University of Tokyo()
Date 2006-03-16
Paper # PRMU2005-258
Volume (vol) vol.105
Number (no) 673
Page pp.pp.-
#Pages 2
Date of Issue