Real-time Face Detection using FFS boosting method in Hierarchical Feature Spaces

摘要：

AdaBoost based training method hasbecome a state-of-the-art boosting approach in facedetection system. In this paper, compared to thenaive AdaBoost method, Forward Feature Selection(FFS) method is used in feature selection to reducethe training time by about 50 to 100 times withoutloss of performance. Furthermore, hierarchicalfeature spaces (both local and global) to construct adetector cascade based on FFS method are adopted,which still have good discrimination in the later stageof boosting process. Experimental results show thatour method can achieve higher performance usingfar less training time.

作者: Hao Ji Fei Su Feng Ye Yuanbo Chen Yujia Zhu

作者单位: School of Information and Telecommunications Beijing University of Posts and Telecommunications Beij School of Electronic Engineering Beijing University of Posts and Telecommunications Beijing 100876,

会议类型: 国际会议

会议名称: The 4th International Conference on Bioinformatics and Biomedical Engineering(第四届IEEE生物信息与生物医学工程国际会议 iCBBE 2010)

会议地点: 成都

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-18（万方平台首次上网日期，不代表论文的发表时间）