@InProceedings{LeitePeGoVeSaNaCa:2007:LeEyDe,
author = "Leite, Bruno de Brito and Pereira, Eanes Torres and Gomes, Herman
Martins and Veloso, Luciana Ribeiro and Santos, Cī\ı and
Nascimento, cero Einstein do and de Carvalho, Jo{\~a}o Marques",
affiliation = "Departamento de Sistemas e Computa{\c{c}}{\~a}o, Universidade
Federal de Campina Grande and Departamento de Sistemas e
Computa{\c{c}}{\~a}o, Universidade Federal de Campina Grande and
Departamento de Sistemas e Computa{\c{c}}{\~a}o, Universidade
Federal de Campina Grande and Departamento de Engenharia
El{\'e}trica, Universidade Federal de Campina Grande and
Departamento de Engenharia El{\'e}trica, Universidade Federal de
Campina Grande and Departamento de Engenharia El{\'e}trica,
Universidade Federal de Campina Grande",
title = "A learning-based eye detector coupled with eye candidate filtering
and PCA features",
booktitle = "Proceedings...",
year = "2007",
editor = "Falc{\~a}o, Alexandre Xavier and Lopes, H{\'e}lio C{\^o}rtes
Vieira",
organization = "Brazilian Symposium on Computer Graphics and Image Processing, 20.
(SIBGRAPI)",
publisher = "IEEE Computer Society",
address = "Los Alamitos",
keywords = "eye detection, neural networks, principal component analysis,
integral image.",
abstract = "In this work, we present a system based on a Neural Network
classifier for eye detection in human face images. This classifier
works on eye candidate regions extracted from a face image and
represented by a reduced number of features, selected by Principal
Component Analysis. The regions are determined considering that in
an image window containing the eye, the gray level distribution
will generally assume a pattern of adjacent light-dark-light
horizontal and vertical stripes, corresponding to the eyelid,
pupil and eyelid, respectively. For training, validation and
testing, a database was built with a total of 4,400 images.
Experimental results have shown that the proposed approach
correctly detects more eyes than any of two existing systems
(Rowley-Baluja-Kanade and Machine Perception Toolbox), for eye
location error tolerances from 0 to 5 pixels. Considering an error
tolerance of 9 pixels, the correct detection rate achieved was
above 90%. .",
conference-location = "Belo Horizonte, MG, Brazil",
conference-year = "7-10 Oct. 2007",
doi = "10.1109/SIBGRAPI.2007.44",
url = "http://dx.doi.org/10.1109/SIBGRAPI.2007.44",
language = "en",
ibi = "6qtX3pFwXQZG2LgkFdY/R2tfw",
url = "http://urlib.net/ibi/6qtX3pFwXQZG2LgkFdY/R2tfw",
targetfile = "gomes-EyeDetection.pdf",
urlaccessdate = "2025, Feb. 09"
}