Interactions gestuelles en réalité augmentée appliquée au cas de l’analyse structurale

Abstract

In recent years, virtual and augmented reality increased in popularity among
both businesses and individuals, thanks to the availability of devices that facilitate
access to this kind of technology. In this perspective, the EFFaTA-MeM
research project notably considers the functionalities offered by augmented reality
in order to design visualization and interaction tools for structural text
analysis. This work aims to explore the Microsoft HoloLens headset capabilities
through the design of visualizations, and their manipulation, to assist analysts.
The results of this exploration are the development of a visualization’s prototype
for which a set of manipulation gestures is defined and requires an extension
of gesture detection and recognition capabilities of the HoloLens, which can be
implemented using two proposed software architectures. In addition, a proof of
concept using a Leap Motion hand detection device is performed and detailed in
this document. Finally, the presented results pave the way for future work such
as the exploration of other paradigms of interactions or the study of machine
learning methods dedicated to the definition of gestures to ease their recognition
process.

Date of Award	19 Jun 2018
Original language	French
Awarding Institution	University of Namur
Supervisor	Bruno Dumas (Supervisor) & Anne Wallemacq (Co-Supervisor)