A Regularized Approach for Geodesic-Based Semisupervised Multimanifold   Learning

doi:10.1109/TIP.2014.2312643

CORC > 北京大学 > 信息科学技术学院

	A Regularized Approach for Geodesic-Based Semisupervised Multimanifold Learning
	Fan, Mingyu ; Zhang, Xiaoqin ; Lin, Zhouchen ; Zhang, Zhongfei ; Bao, Hujun
刊名	ieee transactions on image processing
	2014
关键词	Feature extraction manifold learning semisupervised learning image classification NONLINEAR DIMENSIONALITY REDUCTION FACE RECOGNITION FEATURE-EXTRACTION DISCRIMINANT-ANALYSIS PRESERVING PROJECTIONS GEOMETRIC FRAMEWORK UNIFIED FRAMEWORK MANIFOLD SUBSPACE ISOMAP
DOI	10.1109/TIP.2014.2312643
英文摘要	Geodesic distance, as an essential measurement for data dissimilarity, has been successfully used in manifold learning. However, most geodesic distance-based manifold learning algorithms have two limitations when applied to classification: 1) class information is rarely used in computing the geodesic distances between data points on manifolds and 2) little attention has been paid to building an explicit dimension reduction mapping for extracting the discriminative information hidden in the geodesic distances. In this paper, we regard geodesic distance as a kind of kernel, which maps data from linearly inseparable space to linear separable distance space. In doing this, a new semisupervised manifold learning algorithm, namely regularized geodesic feature learning algorithm, is proposed. The method consists of three techniques: a semisupervised graph construction method, replacement of original data points with feature vectors which are built by geodesic distances, and a new semisupervised dimension reduction method for feature vectors. Experiments on the MNIST, USPS handwritten digit data sets, MIT CBCL face versus nonface data set, and an intelligent traffic data set show the effectiveness of the proposed algorithm.; Computer Science, Artificial Intelligence; Engineering, Electrical & Electronic; SCI(E); EI; 1; ARTICLE; fanmingyu@amss.ac.cn; zhangxiaoqinnan@gmail.com; zlin@pku.edu.cn; zhongfei@cs.binghamton.edu; bao@cad.zju.edu.cn; 5; 2133-2147; 23
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/213677]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Fan, Mingyu,Zhang, Xiaoqin,Lin, Zhouchen,et al. A Regularized Approach for Geodesic-Based Semisupervised Multimanifold Learning[J]. ieee transactions on image processing,2014.
APA	Fan, Mingyu,Zhang, Xiaoqin,Lin, Zhouchen,Zhang, Zhongfei,&Bao, Hujun.(2014).A Regularized Approach for Geodesic-Based Semisupervised Multimanifold Learning.ieee transactions on image processing.
MLA	Fan, Mingyu,et al."A Regularized Approach for Geodesic-Based Semisupervised Multimanifold Learning".ieee transactions on image processing (2014).