Title :
An Improved Algorithm for Diverse AdaBoostSVM
Author :
Guo, Song ; Gu, Guochang ; Liu, Haibo ; Shen, Jing ; Li, Changyou
Author_Institution :
Sch. of Comput. Sci. & Technol., Harbin Eng. Univ., Harbin, China
Abstract :
In order to improve the training convergence speed and detection accuracy of diverse AdaBoostSVM, an improved algorithm is proposed according to the asymmetry in face detection. In the algorithm, the weight of each weak learner, which represents importance of each weak learner, is determined by the error rate and the recognition capability of the weak learner for the face samples. The results of the experiments show that the proposed algorithm could improve the training convergence speed and the detection accuracy in face detection.
Keywords :
convergence; face recognition; image sampling; support vector machines; detection accuracy; diverse AdaBoostSVM; error rate; face detection; face sample; recognition capability; training convergence speed; weak learner; Aerospace engineering; Algorithm design and analysis; Application software; Boosting; Computer science; Convergence; Error analysis; Face detection; Face recognition; Support vector machines;
Conference_Titel :
Computational Sciences and Optimization, 2009. CSO 2009. International Joint Conference on
Conference_Location :
Sanya, Hainan
Print_ISBN :
978-0-7695-3605-7
DOI :
10.1109/CSO.2009.452
