News Release

UCSD wins $600k anti-terrorism award

Grant and Award Announcement

University of California - San Diego

San Diego, May 29, 2003 -- With funding from a federal interagency organization for combating terrorism, scientists at the University of California, San Diego (UCSD) are developing an automated system for detecting and tracking faces in a crowd. The surveillance system links a network of 360-degree cameras which interact "intelligently," thanks to sophisticated computer algorithms developed at UCSD's Computer Vision and Robotics Research (CVRR) laboratory.

The Technical Support Working Group (TSWG) has awarded $600,000 to the 18-month project led by CVRR director Mohan Trivedi, a professor of electrical and computer engineering at UCSD's Jacobs School of Engineering. "This award will allow us to build on some of the work we are doing with agencies such as SPAWAR and the Office of Naval Research," said Trivedi, who is also a program leader in the California Institute for Telecommunications and Information Technology [Cal-(IT)2], a partnership of UCSD and UC Irvine . "Our close relationships with corporations and local agencies in San Diego, including 'first responders,' will also allow us to test these new systems very quickly in the real world once the technology is developed."

TSWG is a twenty-year old interagency forum that identifies, prioritizes, and coordinates interagency and international research and development (R&D) requirements for combating terrorism. In response to a special Broad Agency Announcement (a call for bids) post-9/11, TSWG received over 12,500 research proposals from industry, academic, government and national laboratories. Of those, only roughly 60 have been funded, to the tune of approximately $50 million.

The UCSD project is one of nine winning proposals that deal with video for public safety and surveillance. "Detecting and tracking a possible terrorist is of paramount importance in combating terrorism" said program manager David Herrington. "UCSD's proposal was consistent with our overall view that networks of 'smart' cameras could be one of the principal solutions to this problem -- and a solution that could happen rather quickly with the appropriate level of funding."

The UCSD researchers have already developed multi-camera systems for monitoring traffic. Dubbed DIVAs (for 'distributed interactive video arrays'), they link different types of cameras, including omnivision, 'panoramic' cameras. "The TSWG project is a great recognition for UCSD and it is clearly an important project for us," said Trivedi. "It gives us an opportunity to contribute towards making our communities safer. It also involves resolution of some challenging research problems in multi-camera, system-based tracking and event recognition. A few of the same challenges present themselves when we switch from watching traffic on a highway, to scanning faces in a crowd." One example Trivedi points out:: illumination -- lighting: "These cameras need to be able to detect a face or body whether it's a dark cloudy day, or a sunny one. So we are developing computer algorithms that will allow the cameras to automatically adjust for any lighting level." Another important feature of the UCSD project is to incorporate automatic camera "hand-over" capability, to make multiple cameras work cooperatively. Based upon the tracking results performed by one set of cameras, other cameras can capture higher resolution images for human and event recognition.

Trivedi says another research goal in the future is to make the system reconfigurable. "If a crisis happens in an area not currently covered by surveillance, a mobile network could be dropped into the crisis area, and quickly reconfigure itself to adapt to the new location," he explained. "We need to find ways for optimizing power, bandwidth, quality of service for networks of video and other sensor networks" The effort will include development of wireless networking for the outdoor cameras -- tapping a key area of expertise at both the Jacobs School and Cal-(IT)2.

Originally set up in 1982, TSWG is a stand-alone interagency working group under the oversight of the Department of State's Coordinator for Counterterrorism, and is managed by the Department of Defense. TSWG's total funding has risen from $8 million in fiscal 1992 to $111 million in FY 2002, and roughly $200 million in the current fiscal year. Most of its funding now comes from its 80 participating departments and agencies, for which it acts as a clearinghouse for investments in technology for combating terrorism, with a focus on shorter-term projects (typically 12-18 months) than are typical for other defense-related agencies such as the Defense Advanced Research Projects Agency (DARPA).

###

Note to Online Editors: Websites are welcome to post direct links to two video clips of Professor Trivedi about his research (RealPlayer only).
- On his research objectives
Length: 1:17
http://rpvss.ucsd.edu:8080/ramgen/jsoe/TrivediTSWG.rm

- On why UCSD is uniquely suited to do this research.
Length: 1:39
http://rpvss.ucsd.edu:8080/ramgen/jsoe/TrivediTSWG2.rm

Related Links:

Technical Support Working Group http://www.tswg.gov/tswg/home.htm
Computer Vision and Robotics Research Lab http://cvrr.ucsd.edu
Cal-(IT)2 http://www.calit2.net
Jacobs School of Engineering http://www.jacobsschool.ucsd.edu/


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.