A survey on 2d object tracking in digital video

• Main Authors: Khobragade, A.S., Dethe, C.G., Kulat, K.D.
• Format: Conference or Workshop Item
• Language: English
• Published: 2009
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/13480/1/PID66.pdf; http://repo.uum.edu.my/13480/; http://www.icoci.cms.net.my

This paper presents object tracking methods in video.Different algorithms based on rigid, non rigid and articulated object tracking are studied. The goal of this article is to review the state-of-the-art tracking methods, classify them into different categories, and identify new trends.It is often the case that tracking objects in consecutive frames is supported by a prediction scheme. Based on information extracted from previous frames and any high level information that can be obtained, the state (location) of the object is predicted.An excellent framework for prediction is kalman filter, which additionally estimates prediction error.In complex scenes, instead of single hypothesis, multiple hypotheses using Particle filter can be used.Different techniques are given for different types of constraints in video.