Perspective invariant angle ordering

David Shaw*, Nick Barnes

*Corresponding author for this work

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    In this paper we present the geometric property of perspective invariant angle ordering; the order of angles between point features. We describe how this can be used to exploit the structure of the appearance of features on planar or near planar surfaces to improve precision for localisation and object recognition. We show test results on real-world images that show marked improvement over straight bag-of-features approaches.

    Original languageEnglish
    Title of host publicationDICTA 2009 - Digital Image Computing
    Subtitle of host publicationTechniques and Applications
    Pages256-263
    Number of pages8
    DOIs
    Publication statusPublished - 2009
    EventDigital Image Computing: Techniques and Applications, DICTA 2009 - Melbourne, VIC, Australia
    Duration: 1 Dec 20093 Dec 2009

    Publication series

    NameDICTA 2009 - Digital Image Computing: Techniques and Applications

    Conference

    ConferenceDigital Image Computing: Techniques and Applications, DICTA 2009
    Country/TerritoryAustralia
    CityMelbourne, VIC
    Period1/12/093/12/09

    Fingerprint

    Dive into the research topics of 'Perspective invariant angle ordering'. Together they form a unique fingerprint.

    Cite this