Communications of the ACM
A Brain Boost For Information Overload
A team at Columbia University including Professor Paul Sajda, shown here wearing an electroencephalography cap, are building a computer capable of helping people sift through vast amounts of visual data.
Credit: Eileen Barroso / Columbia University
Columbia University professors Paul Sajda and Shih-Fu Chang have developed C3Vision, a computer vision system that they say could revolutionize how huge amounts of visual data are processed by using a computer to increase the power of the human brain. C3Vision combines the object recognition abilities of humans with the massive processing capabilities of computers, Sajda says.
The C3Vision system features a device that monitors brain activity as a user views a database of photographs, and uses electroencephalography data to rank which pictures created the strongest cortical recognition responses. The system identifies patterns in the visual characteristics of different high-ranking photos, such as color, texture, and shapes. It then scans the database of more than 50 million images and locates those that rank high in the important visual characteristics.
The U.S. Defense Advanced Research Projects Agency has provided $2.4 million to develop an 18-month testing program for the technology and the U.S. National Geospatial-Intelligence Agency wants to use C3Vision to search for objects in satellite images.
From Columbia University
View Full Article
Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA 
No entries found