search-head

Sie sind hier: Homepage >  Suche

    Deep learning based keypoint rejection system for underwater visual ego-motion estimation

    Deep learning based keypoint rejection system for underwater visual ego-motion estimation

    Freier Zugriff

    Most visual odometry (VO) and visual simultaneous localization and mapping (VSLAM) systems rely heavily on robust keypoint detection and matching. With regards to images taken in the underwater environment, phenomena like shallow water caustics and/or dynamic objects like fishes can lead to the detection and matching of unreliable (unsuitable) keypoints within the visual motion estimation pipeline. We propose a plug-and-play keypoint rejection system that rejects keypoints unsuitable for tracking in order to obtain a robust visual ego-motion estimation. A convolutional neural network is trained in a supervised manner, with image patches having a detected keypoint in its center as input and the probability of such a keypoint suitable for tracking and mapping as output. We provide experimental evidence that the system prevents to track unsuitable keypoints in a state-of-the-art VSLAM system. In addition we evaluated several strategies aimed at increasing the inference speed of the network for real-time operations. ; publishedVersion

    Zugriff

    Download

    Exportieren, teilen und zitieren

    Dokumentinformationen

    Titel :

    Deep learning based keypoint rejection system for underwater visual ego-motion estimation

    Beteiligte:
    Leonardi, Marco (Autor:in) / Fiori, Luca (Autor:in) / Stahl, Annette (Autor:in)
    Erscheinungsdatum :

    2020-01-01

    Anmerkungen:

    cristin:1831152
    IFAC-PapersOnLine

    DOI :

    https://doi.org/10.1016/j.ifacol.2020.12.2420

    Medientyp :

    Aufsatz (Zeitschrift)

    Format :

    Elektronische Ressource

    Sprache :

    Englisch

    Schlagwörter :

    Robotsyn , Robotic Vision , Undervannsrobotikk , Underwater robotics , Robotikk , Robotics , Maskinlæring , Machine learning

    Klassifikation :

    DDC:    629

    Lizenzbestimmungen :

    http://creativecommons.org/licenses/by-nc-nd/4.0/deed.no

    Ähnliche Titel

    SELF-SUPERVISED 3D KEYPOINT LEARNING FOR EGO-MOTION ESTIMATION

    TANG JIEXIONG / AMBRUS RARES A / GUIZILINI VITOR et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Self-supervised 3D keypoint learning for ego-motion estimation

    TANG JIEXIONG / AMBRUS RARES A / GUIZILINI VITOR et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    SELF-SUPERVISED 3D KEYPOINT LEARNING FOR MONOCULAR VISUAL ODOMETRY

    TANG JIEXIONG / AMBRUS RARES A / GUIZILINI VITOR et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Visual object categorization with new keypoint-based adaBoost features

    Bdiri, Taoufik / Moutarde, Fabien / Steux, Bruno | IEEE | 2009

    Visual Object Categorization with New Keypoint-based AdaBoost Features

    Bdiri, T. / Moutarde, F. / Steux, B. | British Library Conference Proceedings | 2009

    Zugriff

    Download

    Seitennavigation

    Dokumentinformationen

    Ähnliche Titel

    Exportieren, teilen und zitieren