A tutorial on Principal Components Analysis 原著:Lindsay I Smith, A tutorial on Principal Components Analysis, February 26, 2002. 翻译:houchaoqun.时间:2017/01/18.出处:http://blog.csdn.net/houchaoqun_xmu | http://blog.csdn.net/Houchaoqun_XMU/article/details
<Aggregating local descriptors into a compact image representation>论文笔记 在论文中,提取到VLAD特征后,要对特征向量进行PCA降维,就是用一个大小为D' * D的矩阵M,对VLAD特征向量x做变换,降维后的vector是x' = Mx,x'的大小是D'维.矩阵M是由原样本的协方差矩阵的D'个特征向量构成. 为什么M要是特征向量的矩阵呢? 根据PRML中的内容,理解如下: 1,Maxinum Variance Formula
http://matlabdatamining.blogspot.com/2010/02/principal-components-analysis.html 英文Principal Components Analysis的博客,写的挺好,担心以后打不开,全文转载. Principal Components Analysis Introduction Real-world data sets usually exhibit relationships among their variables.