PROCEDURE TO DESCRIBE THE GEOMETRIC BEHAVIOR OF HUMANS ON A SCENE CATCHED BY AN ARTIFICIAL VISION SYSTEM, BASED ON A MODEL OF BLOCKS AND, SPECIAL, ORIENTED TO THE VIDEO-SURVEILLANCE TASK.

摘要

Procedure to describe the behaviorgeometric of humans in a scene captured by a vision systemartificial, based on a block model and, especially, orientedto the task of video surveillance.;Procedure that is able to monitor, inreal time and with a minimum computational load, the"geometric behavior" of a human in a scene capturedfor an artificial vision system. Where, for "behaviorgeometric "of a human, we refer to the description ofevents in a 3D space related to static characteristics(position) and dynamics (movement) of a human's behavioror parts of it.;The procedure is based on characterizing thehuman using a block model. This model, relativelysimple, it allows to treat similarly different views of the humanand provides the necessary information for a robust descriptionand reliable of different events of interest for the task ofartificial vision surveillance.