The Community for Technology Leaders
Pacific-Asia Workshop on Computational Intelligence and Industrial Application, IEEE (2008)
Dec. 19, 2008 to Dec. 20, 2008
ISBN: 978-0-7695-3490-9
pp: 376-380
ABSTRACT
In this paper, a robust and effective face detection method with HTF-Boosting is proposed. Firstly, a new feature, called Haar texture feature, is proposed that has many merits compared with Haar-Like feature. Secondly, a new Boosting algorithm, called Haar Texture Feature Boosting (HTF-Boosting), is proposed to construct strong face/nonface classifiers. The HTF-Boosting algorithm trains strong classifiers with with a smaller number of weak classifiers and a little time. What is more, HTF-Boosting algorithm yields higher classification accuracy than AdaBoost algorithm using Haar-Like feature.The experimental results on MIT-CBCL dataset demonstrate HTF-Boosting outperforms traditional AdaBoost. Finally, the test results on MIT+CMU frontal face test set show our face detector is more effective than relative detector. In addition, the proposed algorithm is successfully applied to real-time detection of face and eyes state during driving.
INDEX TERMS
AdaBoost, HTF-Boosting, Face Detection
CITATION
Chunxia Zhao, Yunyang Yan, Jingyu Yang, Zhibo Guo, "HTF-Boosting Learning and Face Detection", Pacific-Asia Workshop on Computational Intelligence and Industrial Application, IEEE, vol. 01, no. , pp. 376-380, 2008, doi:10.1109/PACIIA.2008.83
93 ms
(Ver 3.3 (11022016))