Robust automated planar normalization of tracking data

This paper presents an automated method for normalizing measurable properties of tracked objects that lie primarily on a ground plane in far-field tracking scenarios (e.g., video surveillance). The normalization of properties such as sizes, lengths, and heights is accomplished automatically and reliably without assumptions on the type of objects or camera geometry. This normalization enables comparison of properties of tracked objects in different areas of the scene, estimation of the actual property values of all tracked objects with the specification of a single tracked object’s property, and comparison of object properties across different scenes with the specification of a single relative parameter. Results are shown for different properties of realistic tracking data from a variety of scenes.