Presentation 2007-02-23
Calibration of Boosted Cascade Classifier and Distribution-based Face Detection
Hiromasa TAKATSUKA, Masayuki TANAKA, Masatoshi OKUTOMI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Face detection is a useful technique in computer vision. Many researches have been developed in the literature. These existing studys have focused on subwindows and could not integrate the information over image space and scale in their framework. We found the classifier value around the interesting points keep high on the true face and have proposed the distribution-based face detection framework in a previous report. In this report, we propose a distribution-based method using the boosted cascade classifier. The output of the boosted cascade classifier is independent between each stage. We propose the novel calibration method for the boosted cascade classifier. Finally, we evaluate the local face likelihood distribuiton over space and scale by distribution-based framework. Experimental results using 170 input scenes, which might or might not include faces, show that the proposed framework improves the detection rate by about 5%.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Calibration / AdaBoost Classifier / Face Detection
Paper # PRMU2006-230,HIP2006-123
Date of Issue

Conference Information
Committee HIP
Conference Date 2007/2/16(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) Calibration of Boosted Cascade Classifier and Distribution-based Face Detection
Sub Title (in English)
Keyword(1) Calibration
Keyword(2) AdaBoost Classifier
Keyword(3) Face Detection
1st Author's Name Hiromasa TAKATSUKA
1st Author's Affiliation Department of Mechanical and Control Engineering, Tokyo Institure of Technology()
2nd Author's Name Masayuki TANAKA
2nd Author's Affiliation Department of Mechanical and Control Engineering, Tokyo Institure of Technology
3rd Author's Name Masatoshi OKUTOMI
3rd Author's Affiliation Department of Mechanical and Control Engineering, Tokyo Institure of Technology
Date 2007-02-23
Paper # PRMU2006-230,HIP2006-123
Volume (vol) vol.106
Number (no) 541
Page pp.pp.-
#Pages 6
Date of Issue