Current state-of-the-art object detection algorithms still suffer the problem of imbalanced distribution of training data over object classes and background. Recent work introduced a new loss function called focal loss to mitigate this problem, but at the cost of an additional hyperparameter. Manually tuning this hyperparameter for each training task is highly time-consuming. With automated focal loss we introduce a new loss function which substitutes this hyperparameter by a parameter that is automatically adapted during the training progress and controls the amount of focusing on hard training examples. We show on the COCO benchmark that this leads to an up to 30% faster training convergence. We further introduced a focal regression loss which on the more challenging task of 3D vehicle detection outperforms other loss functions by up to 1.8 AOS and can be used as a value range independent metric for regression.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Automated Focal Loss for Image based Object Detection


    Beteiligte:
    Weber, Michael (Autor:in) / Furst, Michael (Autor:in) / Zollner, J. Marius (Autor:in)


    Erscheinungsdatum :

    2020-10-19


    Format / Umfang :

    345785 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Astronomical Image Analyzer with Automated Object Detection

    Orellana, Sonny / Nguyen, Richard / Boussalis, Helen et al. | AIAA | 2005


    OPTICAL OBJECT TRACKING ON FOCAL PLANE WITH DYNAMIC FOCAL LENGTH

    DAS SHOUVIK / GUPTA NIKHIL | Europäisches Patentamt | 2021

    Freier Zugriff

    Optical object tracking on focal plane with dynamic focal length

    DAS SHOUVIK / GUPTA NIKHIL | Europäisches Patentamt | 2022

    Freier Zugriff

    Automated Guided Vehicles Using Multiple Focal Points

    SINWOONG SON / YOUNGHOON SHIN | Europäisches Patentamt | 2024

    Freier Zugriff