MPEG-4 Video and Audio Information Processing in Audio Visual MobileRobot Systems

摘要

The popular cases of processing in audio and video robot systems are audio visual speaking person tracking and sound localization. The MPEG-4 standard described and codes video information in form of the objects, which can be the speaking persons observed with mobile robot video camera. It is proposed and tested in this article a method of using video objects and also audio information, from the coded MPEG-4 bit stream for audio visual speaking person tracking and sound localization.