Date of Award

12-2016

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Engineering and Sciences

First Advisor

Anthony O. Smith

Second Advisor

Adrian M. Peter

Third Advisor

Gerogios Anagnostopoulos

Fourth Advisor

Samuel P. Kozaitis

Abstract

Video analysis is a rich research topic, due to the wide spectrum of applications such as surveillance, activity recognition, security, and event detection. One of the important challenges in video analysis is object tracking, which provides the ability to determine the exact location of an object of interest within each frame. Many challenges affect the efficiency of a tracking algorithm such as scene illumination change, occlusion, scaling change and determining a search window from which to track object(s). We present an integrated probabilistic model for object track- ing, that combines implicit dynamic shape representations and probabilistic object modeling. We demonstrate the proposed tracking algorithm on a benchmark video tracking data set, and achieve state-of-the art results in both overlap-accuracy and speed.

Download

Share

COinS