For robust object detection on LiDAR data, neural networks have to be trained on diverse datasets that contain many different environmental influences like rain, snow, or fog. To this date, few datasets, with those features, are available while there exist many datasets recorded under perfect weather conditions. Repurposing those datasets by simulating adverse environmental conditions on top of them and training networks with the resulting enhanced datasets, is intended to lead to more robust neural networks. In the following we propose models to realistically simulate the effects of rain, snow, and fog on LiDAR datasets based on physical and empirical fundamentals. Then we parameterize our simulation to best fit real LiDAR data that was captured in those environments, in order to achieve a highly accurate simulation. Finally, the impact of adverse weather on neural network detection performance is demonstrated.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Simulating Realistic Rain, Snow, and Fog Variations For Comprehensive Performance Characterization of LiDAR Perception


    Contributors:


    Publication date :

    2022-06-01


    Size :

    858995 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Rain and Snow Removal

    Tian, Jiandong | Springer Verlag | 2021


    Vehicle rain and snow removing method, vehicle rain and snow removing system and vehicle with vehicle rain and snow removing system

    HUANG HAITAO / HUANG ZHONGSHUI / WANG YAN et al. | European Patent Office | 2023

    Free access

    RAIN AND SNOW INTRUSION ARRESTER

    KITAMURA TOSHIRO | European Patent Office | 2022

    Free access

    Simulating Photo-realistic Snow and Fog on Existing Images for Enhanced CNN Training and Evaluation

    Bernuth, Alexander von / Volk, Georg / Bringmann, Oliver | IEEE | 2019


    Eis-Snow-Rain-Sensoren (ESRS)

    European Patent Office | 2024

    Free access