This work presents a probabilistic deep neural network that combines LiDAR point clouds and RGB camera images for robust, accurate 3D object detection. We explicitly model uncertainties in the classification and regression tasks, and leverage uncertainties to train the fusion network via a sampling mechanism. We validate our method on three datasets with challenging real-world driving scenarios. Experimental results show that the predicted uncertainties reflect complex environmental uncertainty like difficulties of a human expert to label objects. The results also show that our method consistently improves the Average Precision by up to 7% compared to the baseline method. When sensors are temporally misaligned, the sampling method improves the Average Precision by up to 20%, showing its high robustness against noisy sensor inputs.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Leveraging Uncertainties for Deep Multi-modal Object Detection in Autonomous Driving


    Contributors:
    Feng, Di (author) / Cao, Yifan (author) / Rosenbaum, Lars (author) / Timm, Fabian (author) / Dietmayer, Klaus (author)


    Publication date :

    2020-10-19


    Size :

    1296280 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    LEVERAGING UNCERTAINTIES FOR DEEP MULTI-MODAL OBJECT DETECTION IN AUTONOMOUS DRIVING

    Feng, Di / Cao, Yifan / Rosenbaum, Lars et al. | British Library Conference Proceedings | 2020




    Multi-Modal 3D Object Detection in Autonomous Driving: A Survey and Taxonomy

    Wang, Li / Zhang, Xinyu / Song, Ziying et al. | IEEE | 2023


    MENet: Multi-Modal Mapping Enhancement Network for 3D Object Detection in Autonomous Driving

    Liu, Moyun / Chen, Youping / Xie, Jingming et al. | IEEE | 2024