Using Head Position to Enhance Human Interfaces

Current human-computer interface design relies too heavily on a traditional model that casts the human as an operator of a dumb machine. Methods of human-computer interaction based on the keyboard and mouse alone require reevaluation, insofar as they do not foster a personable dialogue between the user and the computer. The ideal interface uses, among other things, knowledge of the position of the user's head as a significant clue to what the computer should do next. This thesis describes a computer program which uses a video camera to identify the position and orientation of the user's head with respect to a computer screen. Several application programs are developed to demonstrate the consequences of adding this knowledge to a traditional human interface. Thesis Supervisor: Walter Bender Title: Principle Research Scientist, Electronic Publishing Section This work was supported in part by International Business Machines Corporation.