Adaptive information density for augmented reality displays

Augmented Reality (AR) browsers show geo-referenced data in the current view of a user. When the amount of data grows too large, the display quickly becomes cluttered. Clustering items by spatial and semantic attributes can temporarily alleviate the issue, but is not effective against an increasing...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the Workshop on Future Trends of Distributed Computing Systems s. 83 - 92
Hlavní autoři: Tatzgern, Markus, Orso, Valeria, Kalkofen, Denis, Jacucci, Giulio, Gamberini, Luciano, Schmalstieg, Dieter
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.03.2016
Témata:
Browsers
Clustering algorithms
Clutter
Data visualization
Electronic mail
Semantics
Visualization
ISSN:2375-5334
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Augmented Reality (AR) browsers show geo-referenced data in the current view of a user. When the amount of data grows too large, the display quickly becomes cluttered. Clustering items by spatial and semantic attributes can temporarily alleviate the issue, but is not effective against an increasing amount of data. We present an adaptive information density display for AR that balances the amount of presented information against the potential clutter created by placing items on the screen. We use hierarchical clustering to create a level-of-detail structure, in which nodes closer to the root encompass groups of items, while the leaf nodes contain single items. Our method selects items and groups from different levels of this hierarchy based on user-defined preferences and on the amount of visual clutter caused by placing these items. The number of presented items is adapted during user interaction to avoid clutter. We compare our interface to a conventional AR browser interface in a qualitative user study. Users clearly preferred our interface, because it provided a better overview of the data and allowed for easier comparison. In a second study, we evaluated the effect of different degrees of clustering on search and recall tasks. Users generally made fewer errors, when using our interface for a search task, which indicates that the reduced clutter allowed them to stay focused on finding the relevant items.
ISSN:2375-5334
DOI:10.1109/VR.2016.7504691