TY - GEN
T1 - A method for detecting human face region based on generation and selection of kernel features
AU - Arakawa, Junya
AU - Morooka, Ken'ichi
AU - Kang, Yousun
AU - Nagahashi, Hiroshi
PY - 2004
Y1 - 2004
N2 - Recent researches for detecting face regions from images have paid attention to high dimensional kernel features(KFs), which are obtained by a non-linear transformation of original features extracted from images. A Support Vector Machine(SVM) is one of the most prominent learning algorithms for KFs. However, SVM is time-consuming because of needing a large number of KFs to improve the accuracy of the classification. This paper proposes a new method that constructs a classifier between face and non-face regions by generating and choosing KFs based on Kullback-Leibler Divergence(KLD). The KLD means a distance between two distributions of face and non-face data under a given KF, and some KFs of large KLDs are selected for the face detection. Moreover, the use of KLD enables us to generate new KFs, and to deal with different kinds of KFs concurrently. Some experiments show that our method can reduce the number of KFs much more than SVM, and achieve almost equal or better detection rate than that of SVM.
AB - Recent researches for detecting face regions from images have paid attention to high dimensional kernel features(KFs), which are obtained by a non-linear transformation of original features extracted from images. A Support Vector Machine(SVM) is one of the most prominent learning algorithms for KFs. However, SVM is time-consuming because of needing a large number of KFs to improve the accuracy of the classification. This paper proposes a new method that constructs a classifier between face and non-face regions by generating and choosing KFs based on Kullback-Leibler Divergence(KLD). The KLD means a distance between two distributions of face and non-face data under a given KF, and some KFs of large KLDs are selected for the face detection. Moreover, the use of KLD enables us to generate new KFs, and to deal with different kinds of KFs concurrently. Some experiments show that our method can reduce the number of KFs much more than SVM, and achieve almost equal or better detection rate than that of SVM.
KW - Boosting
KW - Face detection
KW - Kullback-Leibler divergence
KW - SVM
UR - http://www.scopus.com/inward/record.url?scp=15744389913&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=15744389913&partnerID=8YFLogxK
U2 - 10.1109/ICSMC.2004.1400653
DO - 10.1109/ICSMC.2004.1400653
M3 - Conference contribution
AN - SCOPUS:15744389913
SN - 0780385667
T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
SP - 2191
EP - 2196
BT - 2004 IEEE International Conference on Systems, Man and Cybernetics, SMC 2004
T2 - 2004 IEEE International Conference on Systems, Man and Cybernetics, SMC 2004
Y2 - 10 October 2004 through 13 October 2004
ER -