Age estimation based on canonical correlation analysis and extreme learning machine

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Age estimation based on canonical correlation analysis and extreme learning machine
	Si, Jie 1; Feng, Jun 1; Bu, Qirong 1; Sun, Xiaohu 1; He, Xiaowei 1; Qiu, Shi2
	2015
会议名称	10th chinese conference on biometric recognition, ccbr 2015
会议日期	2015-11-13
会议地点	tianjin, china
通讯作者	feng, jun (fengjun@nwu.edu.cn)
英文摘要	we proposed a novel age estimation scheme based on feature fusion according to canonical correlation analysis. specifically, the shape and texture attributes of feature points in human faces are characterized by both active appearance model (aam) and local binary pattern (lbp).then, the canonical projective vectors are built via canonical correlation analysis for feature fusion. to improve computational efficiency, we first introduce extreme learning machine (elm) to the field of age estimation, and uncover the relation of the fused features and ground-truth age values for age prediction. the experimental results conducted on fg-net age database show that the proposed method achieves better estimation accuracy while requires less computation time than the state of art algorithms such as bif. © springer international publishing switzerland 2015.
收录类别	EI
产权排序	2
会议录	biometric recognition - 10th chinese conference, ccbr 2015, proceedings
会议录出版地	springer verlag
语种	英语
ISSN号	03029743
内容类型	会议论文
源URL	[http://ir.opt.ac.cn/handle/181661/27714]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.School of Information Science and Technology, Northwest University, Xian, China 2.Xian Institute of Optics and Precision Mechanics of CAS, Xian, China
推荐引用方式 GB/T 7714	Si, Jie,Feng, Jun,Bu, Qirong,et al. Age estimation based on canonical correlation analysis and extreme learning machine[C]. 见:10th chinese conference on biometric recognition, ccbr 2015. tianjin, china. 2015-11-13.