Eye-tracked Gaze Gesture Implementation in an Augmented Reality System

Abstract: Gesturing has long been a way of communicating, and is a large part of human nature. It was only natural that this was brought into our technological development when technology advanced. This thesis aims to expand the known human 'library' of gesturing into that of gestural patterns invoked by visual guidance and eye tracking. With the use of a HoloLens to visualize gestural patterns, in combination with an eye tracker from Tobii to invoke and interact with objects, we try to create a modular gesture pattern system. Testing implicates that the developed application without eye tracking is intuitive, easy to figure out and responsive. There are, however, several improvements that can be made in terms of ergonomic and long term use. Eye tracking implementations to the gaze gesture system is looked into and discussed throughout the thesis, and with further development could be made into a fully viable, modular and intuitive tool for integration of AR with the physical world.