Training accurate object detection models often requires a large amount training data. In some cases, limited imagery, from drastically different perspectives than the desired target view positions and angles, may be available for specific objects of interest. Training accurate models with this imagery may not be possible and require a lot of performance-limiting assumptions. However, it may be possible to use this limited imagery to create a 3D model of the targets and their surrounding area. In this paper, we explore training an object detector using only synthetic imagery to detect rooftop stacks for UAV sampling tasks. We show that this detector performs well on real imagery, and enables autonomous UAV sampling. We also note that this approach is general, and should extend to other objects.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Training Object Detectors with Synthetic Data for Autonomous UAV Sampling Applications


    Contributors:


    Publication date :

    2018-06-01


    Size :

    9165520 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Training Object Detectors With Noisy Data

    Chadwick, Simon / Newman, Paul | IEEE | 2019


    TRAINING CONFIGURATION-AGNOSTIC MACHINE LEARNING MODELS USING SYNTHETIC DATA FOR AUTONOMOUS MACHINE APPLICATIONS

    DEGIRMENCI ALPEREN / HONG WON / BOJARSKI MARIUSZ et al. | European Patent Office | 2023

    Free access

    RefineNet: Refining Object Detectors for Autonomous Driving

    Rajaram, Rakesh Nattoji / Ohn-Bar, Eshed / Trivedi, Mohan Manubhai | IEEE | 2016


    Large-Scale Live Active Learning: Training Object Detectors with Crawled Data and Crowds

    Vijayanarasimhan, S. / Grauman, K. | British Library Online Contents | 2014


    Including Transfer Learning and Synthetic Data in a Training Process of a 2D Object Detector for Autonomous Driving

    Antunes, Miguel / Bergasa, Luis M. / Araluce, Javier et al. | Springer Verlag | 2022