Cornell Department of Computer Science Colloquium
4:15pm, November 1st, 2001
B17 Upson Hall

More Features Than Pixels: A Framework for Constructing Fast and Robust Visual Recognition Algorithms

Paul Viola
Mitsubishi Electric Research Labs (MERL), Cambridge

http://www.ai.mit.edu/people/viola/viola.html

All learning approaches require the acquisition of positive and negative examples, the selection of a suitable feature set, and selection of a learning algorithm. I will demonstrate that a wide range of visual recognition problems can be solved by adopting a huge and varied set of visual features. A variant of AdaBoost is then used to select a small set of features critical for a given task. The AdaBoost procedure provides theoretical bounds on training error, generalization error, and the number of required features. Other theoretical results ensure that the presence of spurious features will not reduce testing error significantly.  We have experimented with as many as 1,000,000,000 binary features on problems with thousands of examples. Though the learning process can take days of computation, the resulting classifiers are extremely efficient and robust. Using this approach we have developed the world's fastest face detection system, which can find faces at 15 frames per second on a laptop. We have also applied similar ideas to problems in image database retrieval. New results include a very fast system for facial analysis, which can locate the eyes and nose, and can guess the gender of the user. I will demonstrate these systems during the talk.

**********************************************************************

Bio:

Before moving to MERL Paul Viola was an Associate Professor of Computer Science and Engineering at the Massachusetts Institute of Technology.  He also spent two years as a visiting scientist in the Computational Neurobiology of the Salk Institute in San Diego.  Paul has a broad background in advanced computational techniques, publishing in the fields of computer vision, neurobiological vision, medical imaging, mobile robotics, machine learning, and automated drug design. Paul was a recipient of a National Science Foundation Career award in 1998. He has worked on research and development with a number of companies including: Compaq, IBM Research, Arris Pharmaceuticals and Intarka.