Colocation for SLAM-Tracked VR Headsets with Hand Tracking

Information

• Publication Type: Journal Paper (without talk)
• Workgroup(s)/Project(s):
  • Virtual Reality Group
• Date: April 2021
• DOI: 10.3390/computers10050058
• ISSN: 2073-431X
• Journal: Computers
• Number: 5
• Open Access: yes
• Volume: 10
• Pages: 1 – 17
• Keywords: colocation, multi-user VR, hand tracking , shared space

Abstract

In colocated multi-user Virtual Reality applications, relative user positions in the virtual environment need to match their relative positions in the physical tracking space. A mismatch between virtual and real relative user positions might lead to harmful events such as physical user collisions. This paper examines three calibration methods that enable colocated Virtual Reality scenarios for SLAM-tracked head-mounted displays without the need for an external tracking system. Two of these methods—fixed-point calibration and marked-based calibration—have been described in previous research; the third method that uses hand tracking capabilities of head-mounted displays is novel. We evaluated the accuracy of these three methods in an experimental procedure with two colocated Oculus Quest devices. The results of the evaluation show that our novel hand tracking-based calibration method provides better accuracy and consistency while at the same time being easy to execute. The paper further discusses the potential of all evaluated calibration methods.

Additional Files and Images

Weblinks

• https://www.mdpi.com/2073-431X/10/5/58
• Entry in reposiTUm (TU Wien Publication Database)
• DOI: 10.3390/computers10050058

BibTeX

@article{reimer-2021-CVR,
  title =      "Colocation for SLAM-Tracked VR Headsets with Hand Tracking",
  author =     "Dennis Reimer and Iana Podkosova and Daniel Scherzer and
               Hannes Kaufmann",
  year =       "2021",
  abstract =   "In colocated multi-user Virtual Reality applications,
               relative user positions in the virtual environment need to
               match their relative positions in the physical tracking
               space. A mismatch between virtual and real relative user
               positions might lead to harmful events such as physical user
               collisions. This paper examines three calibration methods
               that enable colocated Virtual Reality scenarios for
               SLAM-tracked head-mounted displays without the need for an
               external tracking system. Two of these methods—fixed-point
               calibration and marked-based calibration—have been
               described in previous research; the third method that uses
               hand tracking capabilities of head-mounted displays is
               novel. We evaluated the accuracy of these three methods in
               an experimental procedure with two colocated Oculus Quest
               devices. The results of the evaluation show that our novel
               hand tracking-based calibration method provides better
               accuracy and consistency while at the same time being easy
               to execute. The paper further discusses the potential of all
               evaluated calibration methods. ",
  month =      apr,
  doi =        "10.3390/computers10050058",
  issn =       "2073-431X",
  journal =    "Computers",
  number =     "5",
  volume =     "10",
  pages =      "1--17",
  keywords =   " colocation, multi-user VR, hand tracking , shared space",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2021/reimer-2021-CVR/",
}