Presentation 2009-05-29
A Study of Facial Direction Estimation from Low-resolution Images
Masakazu MORIMOTO, Jyunko HIRATA, Kensaku FUJII,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a facial direction estimating scheme from outdoor surveillance camera. To prevent invasions of privacy, we reduce resolutions of surveillance camera images which cannot identify person. The proposed scheme estimates facial direction by using support vector machine (SVM). Some experimental results of facial database images show that, when we use 6x6 pixels facial images as input, we can estimate more than 90% facial directions correctly, which consist of 13 directions with 15 degrees interval. Experimental results of adjusted actual camera images shows that, average estimation error is about 21 degrees.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) SVM / Facial direction / low-resolution image / Surveillance camera
Paper # EA2009-16,SIP2009-16,SP2009-21
Date of Issue

Conference Information
Committee SP
Conference Date 2009/5/21(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 Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Study of Facial Direction Estimation from Low-resolution Images
Sub Title (in English)
Keyword(1) SVM
Keyword(2) Facial direction
Keyword(3) low-resolution image
Keyword(4) Surveillance camera
1st Author's Name Masakazu MORIMOTO
1st Author's Affiliation Graduate School of Engineering, University of Hyogo()
2nd Author's Name Jyunko HIRATA
2nd Author's Affiliation Graduate School of Engineering, University of Hyogo
3rd Author's Name Kensaku FUJII
3rd Author's Affiliation Graduate School of Engineering, University of Hyogo
Date 2009-05-29
Paper # EA2009-16,SIP2009-16,SP2009-21
Volume (vol) vol.109
Number (no) 57
Page pp.pp.-
#Pages 6
Date of Issue