Abstract
We describe a video tracking application using the dual-tree Polar Matching Algorithm. We develop the dynamical and observation models in a probabilistic setting and study the empirical probability distribution of the Polar Matching output. We model the visible and occluded target statistics using Beta distributions. This is incorporated into a Track-Before-Detect (TBD) solution for the overall observation likelihood of each video frame and provides a principled derivation of the observation likelihood. Due to the nonlinear nature of the problem, we design a Rao-Blackwellised Particle Filter (RBPF) for the sequential inference. Computer simulations demonstrate the ability of the algorithm to track a simulated video moving target in an urban environment with complete and partial occlusions.
Highlights
Detection and tracking of a known target in video sequences is a common and important problem in image processing
We focus on the scenario of an unmanned air vehicle (UAV) platform-based image sensor as it attempts to track a ground vehicle traversing a cluttered urban environment
We applied the tracking filter to an UAV video sequence of a vehicle moving in a cluttered urban environment
Summary
Detection and tracking of a known target in video sequences is a common and important problem in image processing. Another reason is that the posterior distribution is likely to be multimodal due to the nature of the video data. It is necessary to develop principled probabilistic models to detect and track the target successfully through the cluttered environment Work such as BraMBLe by Isard and MacCormick [15] demonstrated the flexibility of the particle filter to perform video tracking of people through a static background. A Rao-Blackwellised Particle Filter (RBPF) will be developed that improves the efficiency of the particles simulation through Rao-Blackwellising the visibility variable This is applied to a video sequence to track a vehicle that undergoes occlusions as it moves through a cluttered urban scene.
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.