Augmented reality using spatially multiplexed structured light

Michael Torres, Richard Jassel, Ying Tang

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    1 Scopus citations

    Abstract

    Augmented reality (AR) superimposes computer-generated 3-D graphics or 2-D information on the user's view of a surrounding environment in real-time, enhancing the user's perception of the real world. Typical AR systems either use fiducial markers or information about an object in the scene for AR tracking. The former, called marker-based AR, requires artificial markers be placed in the scene to be tracked by the system. When applied to mobile applications, these markers clutter the small view in the mobile device. Often times in the latter method, markerless AR, there is a requirement for offline training of some sort making it unsuitable for unprepared environments. In applications which do not have these requirements, such as structure from motion (SFM) approaches, there can still be the requirement of an initial reference object in the application. Based on these remarks, this research focuses on markerless AR in mobile applications and proposes a tracking method using structured light (SL). The proposed architecture consists of a three-stage process. In the first stage, invariant feature tracking methods are implemented to track real regions between successive video frames. Spatially multiplexed SL, involving projecting a known active image onto a scene, is used to extract 3D range data from a captured image in the second stage. The final stage utilizes the captured 3D data and tracking information to impose AR activities into the live video.

    Original languageEnglish (US)
    Title of host publication2012 19th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2012
    Pages385-390
    Number of pages6
    StatePublished - Dec 1 2012
    Event2012 19th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2012 - Auckland, New Zealand
    Duration: Nov 28 2012Nov 30 2012

    Publication series

    Name2012 19th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2012

    Other

    Other2012 19th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2012
    Country/TerritoryNew Zealand
    CityAuckland
    Period11/28/1211/30/12

    All Science Journal Classification (ASJC) codes

    • Computer Vision and Pattern Recognition
    • Electrical and Electronic Engineering
    • Mechanical Engineering

    Fingerprint

    Dive into the research topics of 'Augmented reality using spatially multiplexed structured light'. Together they form a unique fingerprint.

    Cite this