Object detection is a fundamental task in autonomous driving. Besides bounding-box-like detection algorithms, uncertainty estimation is necessary for safe and trust-worthy perceptions. Bayesian Neural Networks (BNNs) provide a reliable approach to address the challenge. However, it often becomes computationally prohibitive to apply them to modern large-scale neural networks. This work develops an efficient BNN by combining the Laplace Approximation (LA) with linearized inference. Specifically, we study the effectiveness and computational necessity of a diagonal Hessian approximation in the LA on over-parameterized networks. With numerous quantitative experiments on different types of interference, the proposed method demonstrates the ability for real-time and robust uncertainty description for autonomous driving.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Laplace Approximation for Real-Time Uncertainty Estimation in Object Detection


    Contributors:
    Gui, Ming (author) / Qiu, Tianming (author) / Bauer, Fridolin (author) / Shen, Hao (author)


    Publication date :

    2022-10-08


    Size :

    2603304 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Laplace Approximation for Uncertainty Estimation of Deep Neural Networks

    Humt, Matthias | German Aerospace Center (DLR) | 2019

    Free access

    Uncertainty Estimation in One-Stage Object Detection

    Kraus, Florian / Dietmayer, Klaus | IEEE | 2019



    Object uncertainty detection

    VAN HEUKELOM MATTHEW / LEE TENCIA / WANG KAI ZHENYU | European Patent Office | 2022

    Free access

    Depth-assisted rectification for real-time object detection and pose estimation

    Monte Lima, J. o. | British Library Online Contents | 2016