ETO Guest Speaker Series
Emerging Technology Office Speaker Series
Tuesday, April 28, 2009
11:00 - 12:30, room 6N120, backup room 5N116
VTC to Bedford room 2M148
“Passive Vision and the Power of Collective Imaging”
Robert Pless
Assoc. Prof. of Computer Science and Engineering, Washington University
The web has an enormous collection of live cameras that image parks, roads, cities, beaches, mountains, buildings, parking lots, and so on. Over the last 5 years, I have been working to understand how to effectively use this massively distributed, scalable, and already existing network of cameras. In this talk I will introduce the AMOS (Archive of Many Outdoor Scenes) database, which includes images from 1000 cameras captured every half hour over the last 3 years. I will show preliminary algorithms for geo-locating and calibrating these cameras just from image data, and work designed to annotate the scene and the objects in view.
These algorithms are inspired by a combination of work by time-lapse video artists Jason Salavon and Hiroshi Sugimoto, and the work on natural scene statistics that grounds many bio-mimetic image representations. These algorithms provide a robust basis to analyze time-stamped images from a fixed camera, and I will explore some possibilities in the collective use of all the webcams attached to the Internet as a novel global sensor that may help measure weather, social, and climate changes.
I will also briefly present ongoing research in robust tracking of moving targets from persistent video, and the use of light-field imaging techniques using a novel type of plenoptic camera.
Biography: Robert Pless is an Associate Professor of Computer Science and Engineering at Washington University in St. Louis. He received a B.S. in Computer Science from Cornell University in 1994, and a Ph.D. in Computer Science from the University of Maryland, College Park in 2000. In 2006 he received the NSF CAREER award, and the Emerson Excellence in Teaching award. He has chaired the IEEE Workshop on Omnidirectional Vision and Camera Networks (OMNIVIS) in 2003, the MICCAI Workshop on Manifold Learning and Medical Imaging in 2008, and the upcoming IEEE Workshop on Motion and Video Computing, to be held in December of 2009.
Point-of-contact: Jeff Colombe (Dept. E551), phone x3-5307, jcolombe@mitre.org