Monitoring human behavior in an office environment

The goal of this work is for a computer to recognize the actions of people in an office environment. The system described in this thesis uses context to help recognize what actions people are performing. These actions include entering a room, using a computer terminal, opening a cabinet, picking up a phone, etc. Our system recognizes these actions by using prior knowledge about the layout of the room. The low-level Computer Vision techniques of tracking, skin detection, and scene change detection are used in our system to help perform action recognition. All three of these techniques use color images. The output of this system is both a textual and a key frame description of the recognized actions. The results of this work are promising