We propose a method of improving detection precision (mAP) with the help of the prior knowledge about the scene geometry: we assume the scene to be a plane with objects placed on it. We focus our attention on autonomous robots, so given the robot’s dimensions and the inclination angles of the camera, it is possible to predict the spatial scale for each pixel of the input frame. With slightly modified YOLOv3-tiny we demonstrate that the detection supplemented by the scale channel, further referred as S, outperforms standard RGB-based detection with small computational overhead.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Tiny-YOLO object detection supplemented with geometrical data


    Beteiligte:
    Khokhlov, Ivan (Autor:in) / Davydenko, Egor (Autor:in) / Osokin, Ilya (Autor:in) / Ryakin, Ilya (Autor:in) / Babaev, Azer (Autor:in) / Litvinenko, Vladimir (Autor:in) / Gorbachev, Roman (Autor:in)


    Erscheinungsdatum :

    01.05.2020


    Format / Umfang :

    323666 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Tiny-Object Detection Based on Optimized YOLO-CSQ for Accurate Drone Detection in Wildfire Scenarios

    Tian Luan / Shixiong Zhou / Lifeng Liu et al. | DOAJ | 2024

    Freier Zugriff

    Embedded Real-Time Vehicle and Pedestrian Detection Using a Compressed Tiny YOLO v3 Architecture

    Falaschetti, Laura / Manoni, Lorenzo / Palma, Lorenzo et al. | IEEE | 2024

    Freier Zugriff

    YOLO-ESFM: A multi-scale YOLO algorithm for sea surface object detection

    Wei, Maochun / Chen, Keyu / Yan, Fei et al. | Elsevier | 2025

    Freier Zugriff

    Real Time Object Detection using YOLO Algorithm

    Haritha, I.V.Sai Lakshmi / Harshini, M. / Patil, Shruti et al. | IEEE | 2022


    YOLO Object Detection Algorithm with Hybrid Atrous Convolutional Pyramid

    Wang, Hui / Wang, Zhiqiang / Yu, Lijun et al. | British Library Conference Proceedings | 2022