Digital library of construction informatics and information technology in civil engineering and construction
Digital library
Paper: convr-2013-19
| Paper title: | Performance study on natural marker detection for augmented reality supported facility maintenance |
| Authors: | Koch C,König M,Neges M,Abramovici M |
| Summary: | The operation and maintenance phase is the longest and most expensive life-cycle period of building facilities. Operators need to perform activities to provide a comfortable living and working environment and to upkeep equipment to prevent functionality failures. For that purpose they manually browse, sort and select dispersed and unformatted facility information before actually going on the site. Although some software tools have been introduced they still spent 50% of the on-site work on inspection target localization and navigation. To improve these manual, time consuming and tedious procedures, the authors previously presented a framework that uses BIM-based Augmented Reality (AR) to support facility maintenance tasks. The proposed workflow contains AR supported activities, namely AR-based indoor navigation and AR-based maintenance instructions. An inherent problem of AR is marker definition and detection. As introduced, indoor natural markers such as exit signs, fire extinguisher location signs, and appliances’ labels were identified to be suitable for both navigation and maintenance instructions. However, small markers, changing lighting conditions, low detection frame rates and accuracies might prevent the proposed approach from being practical. In this paper the performance of natural marker detection will be evaluated under different configurations, varying marker types, marker sizes, camera resolutions, and lighting conditions. The detection performance will be measured using a pre-defined metric incorporating detection accuracy, tracking quality, frame rates, and robustness. The result will be a set of recommendations on what configurations are most suitable and practical within the given framework. |
| Type: | paper |
| Year of publication: | 2013 |
| Keywords: | Augmented Reality,Facility Maintenance,Natural Markers,Building Information Modeling,Detection Performance |
| Series: | convr:2013 |
| Download paper: | /pdfs/convr-2013-19.pdf |
| Citation: | Koch C,König M,Neges M,Abramovici M (2013). Performance study on natural marker detection for augmented reality supported facility maintenance. CONVR 2013, http://itc.scix.net/paper/convr-2013-19 |
hosted by University of Ljubljana
includes:
W78
includes:
W78
inspired by SciX, ported by Robert Klinc [2019]