Abstract
In this work, we propose a face detection method based on the Gentle AdaBoost algorithm which is used for construction of binary tree structured strong classifiers. Gentle AdaBoost algorithm update values are constructed by using the difference of the conditional class probabilities for the given value of Haar features proposed by [1]. By using this approach, a classifier which can model image classes that have high degree of in-class variations can be constructed and the number of required Haar features can be reduced.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Viola, P., Jones, M.: Rapid Object Detection Using A Boosted Cascade of Simple Features. In: IEEE Conference on Computer Vision and Pattern Recognition (2001)
Li, S.Z.: Statistical Learning of Multi View Face Detection. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2353, pp. 67–81. Springer, Heidelberg (2002)
Viola, P., Jones, M.: Fast Multi View Face Detection. Technical Report TR2003-96, Mitsubishi Electric Research Laboratories (July 2003)
Wu, B., Ai, H., Huang, C.: Fast Rotation Invariant Multi-View Face Detection based on Real Adaboost. In: AFG 2004, Seoul, Korea, May 17-19 (2004)
Friedman, J., Hastie, T., Tibshirani, R.: Additive Logistic regression: a Statitical View of Boosting. Technical Report, Standford University, August 17 (1998)
Liehart, R., Kuranov, A., Pisarevsky, V.: Empirical Analysis of Detection Cascades of Boosted Classifiers for Rapid Object Detection, MRL Tech report (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Demirkır, C., Sankur, B. (2005). Face Detection Using Look-Up Table Based Gentle AdaBoost. In: Kanade, T., Jain, A., Ratha, N.K. (eds) Audio- and Video-Based Biometric Person Authentication. AVBPA 2005. Lecture Notes in Computer Science, vol 3546. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11527923_35
Download citation
DOI: https://doi.org/10.1007/11527923_35
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-27887-0
Online ISBN: 978-3-540-31638-1
eBook Packages: Computer ScienceComputer Science (R0)