SimultaneousMapping and Localization (SLAM) is amultidisciplinary problem with ramifications within several fields. One of the key aspects for its popularity and success is the data fusion produced by SLAM techniques, providing strong and robust sensory systems even with simple devices, such as webcams in Monocular SLAM. This work studies a novel batch validation algorithm, the highest order hypothesis compatibility test (HOHCT), against one of the most popular approaches, the JCCB. The HOHCT approach has been developed as a way to improve performance of the delayed inverse-depth initialization monocular SLAM, a previously developed monocular SLAM algorithm based on parallax estimation. Both HOHCT and JCCB are extensively tested and compared within a delayed inverse-depth initialization monocular SLAM framework, showing the strengths and costs of this proposal. ; Postprint (published version)


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Parsimonious Real Time Monocular SLAM

    Bresson, G. / Feraud, T. / Aufrere, R. et al. | British Library Conference Proceedings | 2012


    Parsimonious real time monocular SLAM

    Bresson, Guillaume / Feraud, Thomas / Aufrere, Romuald et al. | IEEE | 2012


    Collaborative Monocular SLAM with Crowd Sourced Data

    Huai, Jianzhu / Józków, Grzegorz / Toth, Charles et al. | British Library Conference Proceedings | 2016


    Edge landmarks in monocular SLAM

    Eade, E. / Drummond, T. | British Library Online Contents | 2009


    Towards Robust Single-Shot Monocular SLAM

    Schroeder, Gregory / Hussein, Ahmed | Springer Verlag | 2024