Communications of the ACM
M Research Leads to Improved Human, Object Detection Technology
Computer software developed at the University of Missouri detects humans in a video. The technology creates new possibilities for safety and surveillance.
Credit: University of Missouri
University of Missouri researchers are developing software that would enable computers to search within videos and identify humans and specific objects, as well as perform other video analysis tasks. "The goal of our research is to improve how computers interpret the content of a video and how to identify it," says Missouri professor Tony Han. "There are lots of possibilities with video-based detection, and it could come at quite a low cost compared to object and human detection using other sensors, such as thermal sensors."
Intelligent video surveillance would enable automated systems to quickly call for help if it detects that a human is falling, or make a car stop immediately if it detects a pedestrian in its way, for example. Han and his students are developing algorithms for automatic object detection, and have manually labeled more than 3,000 images with object locations to test their algorithms.
This fall, Han and his students participated in the PASCAL grand challenge in object detection, a contest in which researchers compete to detect objects in 20 categories. Han and his team came in first place for detecting potted plants and chairs and second place for detecting humans, cars, horses, and bikes.
From MU News Bureau (MO)
View Full Article
Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA 
No entries found