Class distance weighted locality preserving projection for automatic age estimation

Kazuya Ueki, Masakazu Miya, Tetsuji Ogawa, Tetsunori Kobayashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

We have developed new dimensionality reduction methods, extended from locality preserving projection (LPP), to estimate age using facial images. LPP seeks a linear transformation matrix such that optimally preserves the neighborhood structure of the data. Our focus has been on expanding LPP by making use of class label information. Specifically, one of our ideas is to assign weights only to the data with close class labels. A local scaling method is used for each class to compute the LPP affinity matrix. Another idea is to assign large weights to two samples with close class labels, i.e., close ages. By doing this, class label information for original data (i.e., age information) can be preserved. We thus call this "class distance weighted linear preserving projection" (CDLPP). Experimental results on a large database showed that CDLPP has more discriminative power than conventional methods such as PCA and LPP.

Original languageEnglish
Title of host publicationBTAS 2008 - IEEE 2nd International Conference on Biometrics
Subtitle of host publicationTheory, Applications and Systems
DOIs
Publication statusPublished - 2008 Dec 1
EventBTAS 2008 - IEEE 2nd International Conference on Biometrics: Theory, Applications and Systems - Arlington, VA, United States
Duration: 2008 Sep 292008 Oct 1

Publication series

NameBTAS 2008 - IEEE 2nd International Conference on Biometrics: Theory, Applications and Systems

Conference

ConferenceBTAS 2008 - IEEE 2nd International Conference on Biometrics: Theory, Applications and Systems
CountryUnited States
CityArlington, VA
Period08/9/2908/10/1

ASJC Scopus subject areas

  • Biotechnology

Fingerprint Dive into the research topics of 'Class distance weighted locality preserving projection for automatic age estimation'. Together they form a unique fingerprint.

Cite this