Abstract #302352

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #302352
Activity Number: 25
Type: Contributed
Date/Time: Sunday, August 3, 2003 : 2:00 PM to 3:50 PM
Sponsor: IMS
Abstract - #302352
Title: Classification Modulo Invariance With Application to Face Recognition
Author(s): Nicolas Hengartner*+ and K. Vixie and B. Fraser and B. Wohlberg
Companies: Los Alamos National Laboratory and Los Alamos National Laboratory and Los Alamos National Laboratory and Los Alamos National Laboratory
Address: 2166 Loma Linda, Los Alamos, NM, 87544-2769,
Keywords: geometric methods ; invariance ; pattern recognition
Abstract:

We introduce new techniques for the analysis of large, high-dimensional data-sets via the construction of appropriate reduced representations that incorporate known geometrical features. We illustrate our ideas in the context of face recognition, for which we developed a classification scheme that incorporates analytic information (tangents and curvature) about invariant manifolds induced from the translation, scaling, and rotation of a face. We think of sets of images that differ only by translation, scaling and rotation as lying on a low-dimensional manifold to which classification should be invariant. The scheme fits within a statistical classification framework utilizing second order statistics of a training set. Using publicly available code from a DARPA funded project at Colorado State University, we have made statistically meaningful performance comparisons with existing techniques on the FERET dataset (a standard set of face images available from NIST); our approach outperforms these by a significant margin.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003