iSEEK - Web

Statistics in Face Recognition: Analyzing Probability Distributions of PCA, ICA and LDA Performance Results Kresimir Delac 1, Mislav Grgic 2 and Sonja Grgic 2 1 Croatian Telecom, Savska 32, Zagreb, Croatia, e-mail: kdelac@ieee.org 2 University of Zagreb, FER, Unska 3/XII, Zagreb, Croatia Abstract In this paper we address the issue of evaluating face

www.face-rec.org/algorithms/Comparisons/delac_01.pdf

Average Rating (0 votes)

PCA vs. ICA: A comparison on the FERET data set

context of a simple, baseline recognition system and the FERET face recognition database. We have tested three different distance metrics – L1 norm, L2 norm, and cosine angle - for both PCA and ICA. We find, contrary to previous reports in the literature, that PCA significantly outperforms ICA when the best...

www2.hawaii.edu/~kyungim/papers/baek_cvprip02.pdf

Average Rating (0 votes)

Back to search results

Total time to process: 944ms - Cache load: 0ms.

Do you want to save this as a favorite page?

iSEEK provides users with a Favorites library that allows them easy access to their most-used websites from any computer. If you have an iSEEK account, the content you just selected can be added to your Favorites page to be revisited any time you want.

If you would like to join the iSEEK community, click the "Register" button below to create your free iSEEK account. The resource you have selected will be added to your new Favorites library after you sign in for the first time.

If you already have an iSEEK account, click the button below to sign in and add the resource to your Favorites library.