Institute for Neural Computation

Machine Perception Lab

Marian Stewart Bartlett, Ph.D.

Institute for Neural Computation
University of California, San Diego, 0523
9500 Gilman Drive
La Jolla, CA 92093-0523

Phone: (858) 822-5241
Fax: (858) 534-2014
Email: marni –at- salk.edu

I am Associate Research Professor at the Institute for Neural Computation, UCSD, and I co-direct the Machine Perception Lab. I study learning in vision, with application to face recognition and expression analysis. One area of focus explores models of unsupervised learning and their application to face processing. For example, principles of information maximization which have been associated with neural coding in early vision, also relate to findings in human face perception, such as other-race effects, face adaptation aftereffects, and perceptual effects relating to typicality. These ideas are reviewed in a recent paper. Another major focus of my research is the development of automatic facial expression recognition systems. In collaboration with other members of the Machine Perception Lab, we have developed an end-to-end system for automatic recognition of a set of basic expressions that works in real-time. We have also made substantial progress toward fully automating the facial action coding system, a comprehensive facial expression description system developed by Paul Ekman and Wallace Friesen. For more information, see the project web page. I received my bachelor's degree in Mathematics and Computer Science from Middlebury College in 1988, and my Ph.D. in Cognitive Science and Psychology from UCSD in 1998. My thesis work was conducted with Terry Sejnowski at the Salk Institute and is described in my book, Face Image Analysis by Unsupervised Learning, Kluwer 2001.

My research interests include

· Image analysis through unsupervised learning.

· Facial identity recognition.

· Facial expression analysis.

· Independent component analysis for pattern recognition.

Publications

Activities:

· Associate Editor for Neurocomputing.

· Participating faculty at the UCSD Science of Learning Center on the Temporal Dynamics of Learning.

· Participating faculty for the IGERT program for Vision and Learning in Humans and Machines at UCSD.

Recent Projects

IJPRAI Special Issue on Face Image Processing and Analysis

Neurocomputing Special Issue on Deveopment and Learning
matlab code for face representations using ICA
Learning in Vision KES 2002
Igert Seminar: CSE 291: Topics on Vision in Humans and Machines

Here are photos of my son, Paul 1 2 3 and daughter, Kate 1 2.

Face Image Analysis

by Unsupervised Learning

Marian Stewart Bartlett

Kluwer International Series on Engineering and Computer Science, V. 612. Boston: Kluwer
Academic Publishers, 2001. Ordering Information (888)-640-7378. Also available on Amazon.com.

Foreword by Terrence J. Sejnowski

Book Jacket

Face Image Analysis by Unsupervised Learning explores adaptive approaches to face image analysis. It draws upon principles of unsupervised learning and information theory to adapt processing to the immediate task environment. In contrast to more traditional approaches to image analysis in which relevant structure is determined in advance and extracted using hand-engineered techniques, [this book] explores methods that have roots in biological vision and/or learn about the image structure directly from the image ensemble. Particular attention is paid to unsupervised learning techniques for encoding the statistical dependencies in the image ensemble.

The first part of this volume reviews unsupervised learning, information theory, independent component analysis, and their relation to biological vision. Next, a face image representation using independent component analysis (ICA) is developed, which is an unsupervised learning technique based on optimal information transfer between neurons. The ICA representation is compared to a number of other face representations including eigenfaces and Gabor wavelets on tasks of identity recognition and expression analysis. Finally, methods for learning features that are robust to changes in viewpoint and lighting are presented. These studies provide evidence that encoding input dependencies through unsupervised learning is an effective strategy for face recognition.

Face Image Analysis by Unsupervised Learning is suitable as a secondary text for a graduate level course, and as a reference for researchers and practioners in industry.

"Marian Bartlett's comparison of ICA with other algorithms on the recognition of facial expressions is perhaps the most thorough analysis we have of the strengths and limits of ICA as a preprocessing stage for pattern recognition."

- T.J. Sejnowski, The Salk Institute

May 7, 2001: