Return to search

Improving 3D Remote Guidance using Shared AR Spaces : Separating responsibility of tracking and rendering 3D AR‐objects / Förbättrande av avståndssamarbete i 3D via delade AR‐rymder

Two common problems in Remote Guidance applications include the remote guides lack of direct control over their view into the worker’s physical environment and the difficulties that arise with trying to place virtual 3D objects in a real 3D environment,via a moving, shaky, 2D image.The first issue can be called a lack of remote spatial awareness, the guide can see only what the worker enables them to see. In the worst case the guide is rendered blind to the task environment while the worker is unable to use their device. A common occurrence is tasks that require both hands.The second issue arises from the inherent difficulty present in trying to correctly place a 3D object using only a limited perspective. Camera shake and unreliable tracking of the physical environment being depicted only further add to this problem. Studies show that 3D annotations make for much more effective means of communication, especially in 3D task environments. Allowing the guide some measure of control over their own view has also been shown to improve the guides ability to aid their partner. This paper investigates a method of Remote Guidance where the task of environment tracking and object placement are separated. A prototype application is developed and tested against a baseline 2D-annotation Remote Guidance tool. The study finds the prototype to be an effective way of placing virtual 3D objects in a remote environment. Experimental results show that communication is indeed made better by the inclusion of 3D objects into Remote Guidance. This comes at the cost of a slight increase in the timetaken to complete a task as the complexity of the 3D tool is greater than the 2D one. Unfortunately, the experiment performed fails to properly account for remote spatial awareness.

Identiferoai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-186884
Date January 2022
CreatorsMansén, Erik
PublisherLinköpings universitet, Institutionen för datavetenskap
Source SetsDiVA Archive at Upsalla University
LanguageEnglish
Detected LanguageEnglish
TypeStudent thesis, info:eu-repo/semantics/bachelorThesis, text
Formatapplication/pdf
Rightsinfo:eu-repo/semantics/openAccess

Page generated in 0.0019 seconds