AUTOMATIC EVENT VIDEOING, TRACKING AND CONTENT GENERATION SYSTEM

摘要

An automatic system (fig. 2, 100) that uses one to three grids (20cm) of overhead cameras (20c) to first video an event area (2), and overall bandwidth is greatly reduced by intelligent hubs (26) thast extract foreground blocks 10m based upon initial and continuously updates background images (2r), and the hub also analyze current image (10c) to constantly locate, classify and track in (3D) the limited number of expected foreground objects (10) and as object (10) of interest are tracked, the system automatically direct ptz perspective view cameras (40c) to follow the activities.