Data augmentation techniques have been focused in recent research as they hold the promise to reduce the need for extensive data acquisition and to enable systematic sampling, e.g., in order to examine underrepresented cases. The question of how and to what extent control over the result is possible and necessary is still open. We propose a novel system for license plate substitution in the wild, which replaces a given license plate within an image crop by another one with a customized pattern. The system is based on a CycleGAN architecture, which respects the plate's pose and dominant image features, such as lighting and image sharpness. Most importantly the system is trained on a set of license plate crops without requiring any label information. We demonstrate the validity of our approach in two sets of experiments: First, a license plate recognition system is trained and evaluated with varying amounts and ratios of synthetic over real-life data, and second, the realism of image features is verified by means of a human acceptance study as well as the Fréchet Inception Distance.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Fully Automated, Realistic License Plate Substitution in Real-Life Images


    Contributors:


    Publication date :

    2021-09-19


    Size :

    2044754 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Automated License Plate System

    VILLALOBOS II CRUZ / VILLALOBOS TINA | European Patent Office | 2019

    Free access


    Real-time Automated License Plate Recognition and Tracking of Runaway Vehicles

    Ramadan, Montaser N. A. / Hilles, Shadi M S / Alkhedher, Mohammad et al. | IEEE | 2022


    LICENSE PLATE SHEET, LICENSE PLATE LAMINATE, AND LICENSE PLATE

    MURAMOTO AKIRA / NAKAJIMA TOSHITAKA | European Patent Office | 2019

    Free access

    Automated processing of video license plate images for traffic data acquisition and analysis

    Shuldiner,P.W. / Woodson,J.B. / D'Agustino,S.A. et al. | Automotive engineering | 1997