Abstract Drones are nowadays considered as a valuable solution to monitor urban traffic. Object detectors face numerous challenges when dealing with high-resolution aerial images captured by drones, due to variations in altitude, viewing angle, and weather conditions. To address these challenges, we present an object detector called Butterfly detector that is tailored to detect objects in aerial images. It is an anchor-free method that leverages field-based representations. We introduce Butterfly fields, a type of composite field that describes the spatial information of output features as well as the scale of the detected objects. We employ a voting mechanism between related Butterfly vectors pointing to the object center. We highlight the benefits of our method for urban traffic monitoring by (i) evaluating the recall/precision rate of our detector on two publicly available drone datasets (UAVDT and VisDrone2019), and (ii) measuring the error rate for flow estimations on our newly released EPFL roundabout dataset. We outperform the performance of previous methods while remaining real-time.

    Highlights Introduce a novel anchor-free object detector called Butterfly Detector. Leverage composite fields to deal with the challenges of detection from aerial images. Improve aerial object detection performance especially in high altitude scenarios.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Traffic perception from aerial images using butterfly fields


    Beteiligte:
    Adaimi, George (Autor:in) / Kreiss, Sven (Autor:in) / Alahi, Alexandre (Autor:in)


    Erscheinungsdatum :

    2023-05-17




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Automatic traffic data collection using aerial video images

    Becker,U.J. / Univ.Karlsruhe,DE | Kraftfahrwesen | 1989


    Automated Traffic Analysis in Aerial Images

    Hößler, Tom / Landgraf, Tom | Fraunhofer Publica | 2014

    Freier Zugriff



    Entity Perception Using Remotely Piloted Aerial Vehicle

    Emilyn, J. Jeba / Chandrika, S. Sri / Susma, T. et al. | Springer Verlag | 2024