Gyroscopic Mouse Input for Wall-Size Displays: Improving interaction using Estimated Point Of Focus
Master Thesis Defence by Anders Markussen
Over the past years, interaction with Wall-Size Displays (WSDs) has become an increasingly popular research topic. Some research have found signi cant advantages in applying aspects of physical navigation to work on WSDs, making input devices that operate without physical restrictions desirable. Many interaction techniques and input devices have been proposed, amongst others, the gyroscopic mouse. The gyroscopic mouse is not bound by the need for at stable surface for operation, making it a potential input device for interaction with physical navigation. However, it also comes with a number of disadvantages, such as low precision and increased fatigue. This thesis describes how user tracking and head pose estimation have been used to minimize the disadvantages of the gyroscopic mouse, while attempting to increase the speed of interaction.
In order to achieve these goals, three main objectives were de ned. (1) Develop a low cost tracking solution that support positioning and head pose estimation of users, (2) to propose and evaluate a number of interaction techniques based on the tracker and a gyroscopic mouse and (3) to use the results from the rst evaluation to propose and evaluate an interaction technique that attempts to improve gyroscopic mouse interaction even further. The thesis describes literature studies and design processes that have been completed. Furthermore, it describes the design and analysis of two empirical studies that have been used to analyse the proposed interaction techniques. The thesis shows that the application of tracking and head pose estimation, made it possible to gain signi cant increases in user satisfaction and signi cant decreases in the amount of fatigue that users experienced compared to regular gyroscopic mouse interaction. Furthermore, it shows that signi cant decreases in Movement Times (MT s) are possible for Estimated Point Of Focus (EPOF) based interaction techniques. Since results have shown that the application of EPOF based techniques can provide increased satisfaction and eciency, they indicate that future research projects on interaction with WSDs could gain advantages by considering the use of EPOF based techniques.
Supervisor: Kasper Hornbæk
Censor: Morten Hertzum, RUC