Super-Resolution of Face Images Using Kernel PCA-Based Prior

TitleSuper-Resolution of Face Images Using Kernel PCA-Based Prior
Publication TypeJournal Articles
Year of Publication2007
AuthorsChakrabarti A, Rajagopalan AN, Chellappa R
JournalMultimedia, IEEE Transactions on
Pagination888 - 892
Date Published2007/06//
ISBN Number1520-9210
Keywordsanalysis;learning-based, analysis;probability;, component, Face, image, method;prior, model;face, principal, probability, recognition;image, reconstruction;image, reconstruction;kernel, resolution;principal, super-resolution;high-resolution

We present a learning-based method to super-resolve face images using a kernel principal component analysis-based prior model. A prior probability is formulated based on the energy lying outside the span of principal components identified in a higher-dimensional feature space. This is used to regularize the reconstruction of the high-resolution image. We demonstrate with experiments that including higher-order correlations results in significant improvements