Extraction of an object model for video tracking
T. Murugas, R. Peplow, Jules‐Raymond Tapamo · 2003
The aim of our research is to develop a fully automated system for segmenting and tracking moving objects. Ultimately, such a system may be incorporated into object-based video compression systems or applications that support content-based functionalities. In this paper, we present a method of spatio-temporal segmentation based on change detection and mathematical morphology. The segmentation results will be used as a model for tracking and extracting the moving object in subsequent frames.