In this paper we describe a Visual Odometry algorithm using a single RGB camera and a low resolution Time-ofFlight (ToF) camera for resolving the scale ambiguity in monocular vision. Two hybrid 3D-to-2D approaches for integrating direct depth measurements from the ToF with the RGB camera are compared. These two methods differ in the association of RGB and ToF camera measurements: one is performed in the 3D space, the other in the image space. The performances of these two approaches are compared using an external measurement system as reference. An important contribution of this paper is that the algorithm is tailored for an inexpensive ToF camera with a very small field-of-view (8.7° × 8.7°) and resolution (8 × 8 pixels). The reason of this choice is that such a ToF camera can be easily integrated in a space oriented measurement system, without a significant increase in mass, dimensions and cost.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Monocular visual odometry aided by a low resolution time of flight camera


    Beteiligte:


    Erscheinungsdatum :

    01.06.2017


    Format / Umfang :

    1700878 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Ground Vehicle Monocular Visual Odometry

    Sabry, Mohamed / Al-Kaff, Abdulla / Hussein, Ahmed et al. | IEEE | 2019


    Monocular Outlier Detection for Visual Odometry

    Buczko, Martin / Willert, Volker | British Library Conference Proceedings | 2017


    Monocular Outlier Detection for Visual Odometry

    Buczko, Martin / Willert, Volker | IEEE | 2017


    Monocular visual odometry for underground railway scenarios

    Etxeberria-Garcia, Mikel / Labayen, Mikel / Eizaguirre, Fernando et al. | British Library Conference Proceedings | 2021


    Robust stereo visual odometry from monocular techniques

    Persson, Mikael / Piccini, Tommaso / Felsberg, Michael et al. | IEEE | 2015