Abstract
Object tracking across non-overlapping views is a hot and important research topic in compute vision. In this paper, a novel method to track the interested object continuously across non-overlapping cameras is presented. This challenging task is taken as two sub-problems: single camera object tracking and object matching across disjoint cameras. An object tracking algorithm which improves Tracking-Learning-Detection (TLD) algorithm by adding background extraction and Kalman filter is presented to deal with the first problem. A new object matching algorithm based on the fusion of global features and local features at the assistance of 3D GIS is also introduced for object matching across disjoint cameras. The proposed approach does not need a training phase and inter-camera calibration. Experiments are carried out on real world videos to validate the proposed approach.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.