Abstract
During the past decade, the cascade AdaBoost approach based on Haar features has been one of the-state-of-art techniques for face detection, due to its good combination of accuracy and efficiency. Nevertheless recent study shows that more than 90 % of Haar features are non-discriminative under the single-threshold weak classifiers. In this paper, we present a new cascade AdaBoost face detection approach based on multi-threshold weak classifiers, called multi-threshold AdaBoost (MTAdaBoost). Compared with the existing multi-threshold approaches, the proposed approach is capable of intelligently choosing multiple thresholds, by computing the solutions of two optimal problems. Furthermore, its computational complexity is O(N), i.e. linear with respect to the number of examples, which is of the same scale as that of the single-threshold weak classifiers. This is made possible by exploiting the Kadane algorithm. The experimental results show that the usage of the proposed multi-threshold weak classifier leads to an effective face detector with much less features and at the same time maintaining the same or even better performance, when comparing to the conventional single-threshold weak classifier. In particular, less features give rise to simpler structure, which amounts to significant reduction of training time as well as faster detection speed. As a result, empirically our method processes an image with only half of the time consumed by the single-threshold weak classifiers.
Similar content being viewed by others
References
Amjad A, Griffiths A, Patwary MN (2012) Multiple face detection algorithm using colour skin modelling. IET Image Process 6(8):1093–1101
Bently J (1984) Programming pears: algorithm design techniques. Commun ACM 27(9):865–873
Brubaker S, Brandt J, Sun J, Mullin M, Rehg J (2008) On the design of cascades of boosted ensembles for face detection. Int J Comput Vis IJCV 77(1–3):65–86
Castrilln-Santana M, Dniz-Surez O, Hernndez-Sosa D, Lorenzo J (2011) A comparison of face and facial feature detectors based on the viola-jones general object detection framework. Mach Vis Appl 22(3):481–494
Cevikalp H, Triggs B, Franc V (2013) Face and landmark detection by using cascade of classifiers. In: Proceedings of the 10th IEEE international conference and workshops on automatic face and gesture recognition (FG), pp 1–7
Freund Y, Schapire RE (1997) A decision-theoretic generalization of on-line learning and an application to boosting. J Comput Syst Sci 55:119–139
Friedman J, Hastie T, Tibshirani R (2000) Additive logistic regression: a statistical view of boosting. Ann Stat 28(2):337–407
Huang C, Ai HZ, Li Y, Lao SH (2005) Vector boosting for rotation invariant multi-view face detection. In: The tenth IEEE international conference on computer vision ICCV, pp 446–453
Huang C, Ai HZ, Li Y, Lao SH (2006) Learning sparse features in granular space for multi-view face detection. In: The 7th international conference on automatic face and gesture recognition, pp 401–406
Jain V, Learned-Miller E (2011) Online domain adaptation of a pretrained cascade of classifiers. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR), pp 577–584
Kawulok M, Szymanek J (2012) Precise multi-level face detector for advanced analysis of facial images. IET Image Process 6(2):95–103
Li S, Zhu L, Zhang Z, Blake A, Zhang H, Shum H (2002) Statistical learning of multi-view face detection. In: Proceedings of the 7th European conference on computer vision ECCV, pp 67–81
Lienhar R, Kuranov A, Pisarevsky V (2003) Empirical analysis of detection cascades of boosted classifiers for rapid object detection. In: Proceedings of the 25th DAGM symposium on pattern recognition, pp 297–304
Liu C, Shum HY (2003) Kullback-leibler boosting. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition, pp 587–594
Overett G, Petersson L (2007) Improved response modelling on weak classifiers for boosting. In: IEEE international conference on robotics and automation, pp 3799–3804
Pham M-T, Cham T-J (2007) Online learning asymmetric boosted classifiers for object detection. In: IEEE conference on computer vision and pattern recognition, pp 1–8
Rasolzadeh B, Petersson L, Pettersson N (2006) Response binning: improved weak classifiers for boosting. In: IEEE Intelligent vehicle symposium, pp 344–349
Schapire RE (1990) The strength of weak learnability. Mach Learn 5(2):197–227
Schapire RE (1999) A brief introduction to boosting. In: Proceedings of the 16th international joint conference on artificial intelligence
Schapire RE, Singer Y (1999) Improved boosting algorithms using confidence-rated predictions. Mach Learn 37(3):297–336
Tsao W-K, Lee AJT, Liu Y-H, Chang T-W, Lin H-H (2010) A data mining approach to face detection. Pattern Recogn 43(3):1039–1049
Viola P, Jones M (2001) Fast and robust classification using asymmetric adaboost and a detector cascade. Adv Neural Inf Process Syst 2(3):1311–1318
Viola P, Jones M (2001) Rapid object detection using a boosted cascade of simple features. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 511–518
Wan C, Tian Y, Chen H, Wang X (2011) Rapid face detection algorithm of color images under complex background. In: Advances in neural networks, ISNN 2011. Lecture notes in computer science, vol 6676, pp 356–363
Wu B, Ai HZ, Huang C, Lao SH (2004) Fast rotation invariant multi-view face detection based on real AdaBoost. In: Proceedings of the sixth IEEE international conference on automatic face and gesture recognition, pp 79–84
Wu JX, Brubaker SC, Mullin MD, Rehg JM (2008) Fast asymmetric learning for cascade face detection. IEEE Trans Pattern Anal Mach Intell 30(3), 369–382
Xiao R, Zhu H, Sun H, Tang X (2007) Dynamic cascades for face detection. In: IEEE 11th international conference on computer vision, pp 1–8
Zhang L, Chu R, Xiang S, Liao S, Li S (2007) Face detection based on multi-block lbp representation. In: Proceedings of the international conference on advances in biometrics, pp 11–18
Zhu X, Ramanan D (2012) Face detection, pose estimation, and landmark localization in the wild. In: Proceedings of IEEE conference on computer vision and pattern recognition (CVPR), pp 2879–2886
Acknowledgements
This work was supported by the National Natural Science Foundation of China under Grant No. 61232013, No. 61271434, No. 61175115. The authors appreciate the valuable suggestions of the reviewers.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, Z., Wang, W., Liu, X. et al. An efficient multi-threshold AdaBoost approach to detecting faces in images. Multimed Tools Appl 74, 885–901 (2015). https://doi.org/10.1007/s11042-013-1703-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-013-1703-6