Synthetic data holds significant potential for improving the efficiency of perception tasks in autonomous driving. This paper proposes a practical data synthesis pipeline that employs multi-agent reinforcement learning (MARL) to automatically generate dynamic traffic participant trajectories and leverages augmented reality (AR) processes to produce photo-realistic images. This AR process blends clean static background images extracted from real photos using image matting techniques, with dynamic foreground images rendered from 3D Computer Aided Design (CAD) models in a rendering engine. We posit that this data synthetic pipe line has strong image photorealism, flexible way of interaction scenarios generation and mature tool chain, which has the prospect of large-scale engineering application.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Synthesizing Data for Autonomous Driving: Multi-Agent Reinforcement Learning Meets Augmented Reality


    Additional title:

    Sae Technical Papers


    Contributors:
    Meng, Chao (author) / Wang, Hanchao (author) / Mei, Jinren (author) / Gu, Kai (author) / Zhang, Song (author) / Wang, Tong (author)

    Conference:

    SAE 2023 Intelligent and Connected Vehicles Symposium ; 2023



    Publication date :

    2023-12-20




    Type of media :

    Conference paper


    Type of material :

    Print


    Language :

    English




    AUTONOMOUS DRIVING VEHICLE AND METHOD FOR PROVIDING AUGMENTED REALITY IN AUTONOMOUS DRIVING VEHICLE

    JI AETTIE / LEE KIHYUNG / JUNG DUKYUNG et al. | European Patent Office | 2021

    Free access

    MontiSim: Agent-Based Simulation for Reinforcement Learning of Autonomous Driving

    Hofer, Tristan / Hoppe, Mattis / Kusmenko, Evgeny et al. | IEEE | 2023


    The augmented cockpit - virtual reality meets automotive

    Bolle,M. / Bosch,Stuttgart,DE | Automotive engineering | 2013


    AUGMENTED REALITY BASED AUTONOMOUS DRIVING SIMULATION METHOD AND SYSTEM

    WON JONG HOON / AHN SANG HUN / IM JIUNG | European Patent Office | 2020

    Free access

    Autonomous Driving with Deep Reinforcement Learning

    Zhu, Yuhua / Technische Universität Dresden | SLUB | 2023