Intelligent Activity Tracking and Perception

Tracking humans in video can have a variety of applications including monitoring shoppers in stores, identifying suspicious activities in public spaces, and ensuring the security of senior home residents. Recent advances in video processing and hardware capability on computers have made it possible to track a single human in a video sequence in real time. However, tracking multiple people at the same time without mixing up the identities of different persons is still a major research challenge. Aidant Intelligent Systems Inc. has already implemented real time tracking algorithms internally. To extend their algorithms to reliably track multiple humans. While we will start by considering the input of a single video camera, we will subsequently introduce multiple high end video cameras and develop new algorithms that can improve the accuracy utilizing multiple videos.

Faculty Supervisor:

Anup Basu

Student:

Chenqiu Zhao

Partner:

AiDANT Intelligent Technology

Discipline:

Computer science

Sector:

Information and communications technologies

University:

Program: