Presentation 2007-11-29
Frontal Face Extraction in Video
Masashi OYA, Makoto KOSUGI, Nobuhiko MUKAI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) As a low-resolution camera is commonly used for monitoring, person's size in an image becomes very small. In this paper, a face detector based on Adaboost with haar-like feature is used for a low-resolution image, and the minimum size of a frontal face that can be extracted is investigated. The minimum size to be detected by this face detector is theoretically decided by the size used for learning, however, this minimum face can't be extracted when the face detector is just applied to an original image. So we show that face extraction not only of the expected size but also of a smaller size than the limit of the detector becomes possible by enlarging the original image.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) image processing / image recognition / low-resolution image / haar-like feature / face detector / face detect
Paper # IE2007-97
Date of Issue

Conference Information
Committee IE
Conference Date 2007/11/22(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 Image Engineering (IE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Frontal Face Extraction in Video
Sub Title (in English)
Keyword(1) image processing
Keyword(2) image recognition
Keyword(3) low-resolution image
Keyword(4) haar-like feature
Keyword(5) face detector
Keyword(6) face detect
1st Author's Name Masashi OYA
1st Author's Affiliation Faculty of Engineering, Musashi Institute of Technology()
2nd Author's Name Makoto KOSUGI
2nd Author's Affiliation Faculty of Engineering, Musashi Institute of Technology
3rd Author's Name Nobuhiko MUKAI
3rd Author's Affiliation Faculty of Engineering, Musashi Institute of Technology
Date 2007-11-29
Paper # IE2007-97
Volume (vol) vol.107
Number (no) 358
Page pp.pp.-
#Pages 6
Date of Issue