Normalization and preimage problem in gaussian kernel PCA
Authors:
- 1 January 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15224880,p. 741-744
Abstract
Kernel PCA has received a lot of attention over the past years and showed usefull for many image processing problems. In this paper we analyse the issue of normalization in Kernel PCA for the pre-image problem. We present a geometric interpretation of the normalization process for the gaussian kernel. As a consequence, we could formulate a correct normalization criterion in centered feature space. Furthermore, we show how the proposed normalization criterion improves previous pre-image methods for the task of image denoising.Keywords
This publication has 5 references indexed in Scilit:
- Connecting the Out-of-Sample and Pre-Image Problems in Kernel Methods2007
- Data Fusion and Multicue Data Matching by Diffusion MapsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2006
- Diffusion maps and coarse-graining: a unified framework for dimensionality reduction, graph partitioning, and data set parameterizationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2006
- The Pre-Image Problem in Kernel MethodsIEEE Transactions on Neural Networks, 2004
- Statistical shape influence in geodesic active contours2002