Presentation 1999/1/22
A method for estimating eye regions from facial images
Tatsuya ODA, Kensaku MORI, Yasuhito SUENAGA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a fast and robust method for estimating eye regions from facial images based on the fact that eye regions are usually darker than other regions. This method extracts eye regions by emphasizing facial images using a minimum filter. Precise eye regions are extracted from the result of eye region extraction using an active contour model. We applied the proposed method to eighty cases of real facial images taken under a variety of environment. The experimental result showed that this method could estimate 74% of eye regions in input images correctly. This method proved promising for estimating eye regons from various facial images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) estimation of eye shape / facial image processing / minimum filter / active contour model
Paper # PRMU98-197
Date of Issue

Conference Information
Committee PRMU
Conference Date 1999/1/22(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A method for estimating eye regions from facial images
Sub Title (in English)
Keyword(1) estimation of eye shape
Keyword(2) facial image processing
Keyword(3) minimum filter
Keyword(4) active contour model
1st Author's Name Tatsuya ODA
1st Author's Affiliation Dept.of Computational Science and Engineering, Graduate School of Engineering, Nagoya University()
2nd Author's Name Kensaku MORI
2nd Author's Affiliation Dept.of Computational Science and Engineering, Graduate School of Engineering, Nagoya University
3rd Author's Name Yasuhito SUENAGA
3rd Author's Affiliation Dept.of Computational Science and Engineering, Graduate School of Engineering, Nagoya University
Date 1999/1/22
Paper # PRMU98-197
Volume (vol) vol.98
Number (no) 528
Page pp.pp.-
#Pages 8
Date of Issue