The problem of using computer vision to track and understand the behavior of human beings is a very important one. It has applications in the areas of human-computer interaction, user interface design, robot learning, and surveillance, among others. At its highest level, this problem addresses recognizing human behavior and understanding intent and motive from observations alone. This is a difficult task, even for humans to perform, and misinterpretations are common. We are studying this problem in the areas of human motion recognition, surveillance, tracking, and activity detection.

In the area of surveillance, automated systems to observe pedestrian traffic areas and detect dangerous action are becoming important. Many such areas currently have surveillance cameras in place. However, all of the image understanding and risk detection is left to human security personnel. This type of observation task is not well suited to humans, as it requires careful concentration over long periods of time. Therefore, there is clear motivation to develop automated, intelligent, vision-based monitoring systems that can aid a human user in the process of risk detection and analysis.

Detection of Abandoned Objects	Detection of Thrown Objects	Detection of Unusual Crowd Activity
Camera Tampering Detection	Perimeter Breach Detection	Detection of Motion in Restricted Areas
Detection of Loitering Individuals

Tracking People with Probabilistic Occlusion Reasoning	Tracking People Using Pan-Tilt-Zoom Cameras	Tracking People in Simultaneous Video Streams

Monitoring Crowded Scenes	View-Dependent Human Motion Recognition	Online Motion Recognition
View-Independent Human Motion Recognition	Monitoring Bus Stops	General Activity Recognition

Learning Static Occlusions from a Moving Figure

Faculty
Dr. Nikolaos Papanikolopoulos

Research Associate
Dr. Osama Masoud

Graduate Students
Stefan Atev
Nathaniel Bird
David Kuo-Wei Hsu
Ajay Joshi
Robert Martin
Evan Ribnick

Undergraduate Students
Tyson Giraud