Active Surveillance in Public Environments

The growing interest in automated surveillance in the context of homeland security, civilian, and military applications is fueling demand for robust and efficient surveillance software. This capability is particularly desirable in public environments such as federal buildings or airports where human...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the International Conference on Image Processing, Computer Vision, and Pattern Recognition (IPCV) p. 1
Main Authors Sweet, Timothy, Nicolescu, Mircea
Format Conference Proceeding
LanguageEnglish
Published Athens The Steering Committee of The World Congress in Computer Science, Computer Engineering and Applied Computing (WorldComp) 01.01.2014
Online AccessGet full text

Cover

More Information
Summary:The growing interest in automated surveillance in the context of homeland security, civilian, and military applications is fueling demand for robust and efficient surveillance software. This capability is particularly desirable in public environments such as federal buildings or airports where human operators have to monitor a large number of camera feeds. Humans have been known to be unreliable and inconsistent in this multitask-oriented environment. In this paper, we present a framework for automatically detecting several types of interactions involving human agents and objects they carry. The proposed system uses a novel combination of vision-based techniques that allow realtime detection, as well as offline searching for instances of events with specific attributes in a large video. We validate the system on a set of sample video sequences exhibiting common events we expect to see in public environments.