# SelectVisAR: Selective Visualisation of Virtual Environments in Augmented Reality

**Authors:** Robbe Cools, Jihae Han, Adalberto L. Simeone  
**arXiv:** [2104.08579](https://arxiv.org/abs/2104.08579)  
**Date:** 2021-04-17  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2104.08579v1/

## Key points

- AR and VR technologies are increasingly popular for entertainment and practical purposes
- SelectVisAR is a selective visualisation system for virtual environments in AR to address potential information surplus
- SelectVisAR enables an AR spectator to perceive a co-present VR user in four distinct visualisation conditions: Interactive, Proximity, Everything, and Dollhouse
- Two variables were investigated in two studies: the level of visual information and the effect of scale
- Participants tended to prefer static visualisations over dynamic ones as they disliked lack of control over what they could see in VEs
- Future work will involve improving these visualisations' generalisability into different contexts while applying them as an interactive simulation where both AR users and VR users interact collaboratively with elements from VEs.

## Summary

The use of augmented reality (AR) and virtual reality (VR) technologies has become increasingly popular in recent years, with many applications emerging for both entertainment and practical purposes. To address the potential surplus of information that the AR user may face due to the two planes of information involved - physical and virtual - researchers have proposed SelectVisAR, a selective visualisation system for virtual environments (VEs) in AR. The SelectVisAR system enables an AR spectator to perceive a co-present VR user in four distinct visualisation conditions: Interactive, Proximity, Everything, and Dollhouse. The design uses a human-centric approach to information filtering by selectively visualising only parts of the VE related to the interactive range and possibilities of the VR user. Two variables were investigated in two studies: the level of visual information and the effect of scale. Regarding scale, smaller visualisations were found to provide a better overview of the VE but had the chance of decoupling users from their tasks. On the effect of scale, it was observed that filtering specific selections of VEs did not significantly affect how well people could identify events in VEs based on interactive range and possibilities. Participants tended to prefer static visualisations over dynamic ones as they disliked lack of control over what they could see in VEs. In terms of selective visualisations' influence on perception, participants identified VE events more accurately using Interactive or Context compared to Everything. While there were no significant differences proving that more selective visualisations improve focus statistically, mean values for competence were marginally higher for selective visualisations in this instance's 'bartender' VE. Future work will involve improving these visualisations' generalisability into different contexts while applying them as an interactive simulation where both AR users and VR users interact collaboratively with elements from VEs. Overall, SelectVisAR's selective visualisation system has potential to improve AR users' perception when interacting with co-present VR users; making it a promising solution for future AR and VR applications.