Acquiring, modelling and improving usability of virtual human bodies

Abstract

This thesis aims to study the inclusion of human body models into Virtual Reality environments and practically solve several problems related to their use. It aims to improve both the visual quality of the avatars by employing suitable methodologies for their acquisition, as well as to develop solutions and further the understanding of the impact of including such avatars into VR Systems. To do this, we provide a comprehensive review of the past work and theory in this area. By deftly combining hardware and software components we present a practical system capable of producing detailed human body representations, including presenting as part of the software pipeline a novel registration technique, \color{black}that allows the registration of patches with little overlap, using a non-rigid-based implicit surface estimation\color{black}. We also present the results of an experiment which investigates the effect of hand length representation on distance compression in VR, thus providing a solution for the negative effects of this spatial incongruence on VR experiences.

Date of Award	1 Nov 2021
Original language	English
Awarding Institution	University of Bath
Supervisor	Yongliang Yang (Supervisor) & Peter Hall (Supervisor)