Special Topics of Gesture Recognition Applied in Intelligent Home Environments

摘要: This report shows how to realize a gesture recognition system for controlling appliances in home environments. It gives brief overview on an existing and clarifies details ergonomic remote control of devices by gestures with the help vision system. The focus is motion detection, object normalization identification, modelling prediction Kalman Filter. A main interest was show through example ARGUS, Filter should be modelled initialized physical human motion. initialization problem based tracking differs from initializing systems, where manuals measurement errors. Most aspects mentioned this were implemented ARGUS prototype.