Presentation 2005-10-28
Robust Human Detection in a Complicated Background using Multiple Gaussian Mixture Skin Models
Rei MOCHIKI, Yuichi UCHIYAMA, Kazuya UEKI, Jiro KATTO, Tetsunori KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a human detection method using multiple skin model. Firstly, we detect skin candidates by applying Gaussian Mixture Models of human skins to the whole image, where the models were achieved for various environments (indoor or outdoor, lighting or no-lighting) in advance. We then select the most suitable model having maximum likelihood for the skin candidates, and carry out human region detection. This paper also shows the effectiveness of the proposed method experimentally.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Gaussian Mixture Model / Human Detection / Skin Model
Paper # PRMU2005-99
Date of Issue

Conference Information
Committee PRMU
Conference Date 2005/10/21(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Robust Human Detection in a Complicated Background using Multiple Gaussian Mixture Skin Models
Sub Title (in English)
Keyword(1) Gaussian Mixture Model
Keyword(2) Human Detection
Keyword(3) Skin Model
1st Author's Name Rei MOCHIKI
1st Author's Affiliation Science and Engineering, Waseda University()
2nd Author's Name Yuichi UCHIYAMA
2nd Author's Affiliation Science and Engineering, Waseda University
3rd Author's Name Kazuya UEKI
3rd Author's Affiliation Science and Engineering, Waseda University
4th Author's Name Jiro KATTO
4th Author's Affiliation /
5th Author's Name Tetsunori KOBAYASHI
5th Author's Affiliation
Date 2005-10-28
Paper # PRMU2005-99
Volume (vol) vol.105
Number (no) 375
Page pp.pp.-
#Pages 6
Date of Issue