This paper presents an end-to-end multi-modal learning approach for monocular Visual-Inertial Odometry (VIO), which is specifically designed to exploit sensor complementarity in the light of sensor degradation scenarios. The proposed network makes use of a multi-head self-attention mechanism that learns multiplicative interactions between multiple streams of information. Another design feature of our approach is the incorporation of the model uncertainty using scalable Laplace Approximation. We evaluate the performance of the proposed approach by comparing it against the end-to-end state-of-the-art methods on the KITTI dataset and show that it achieves superior performance. Importantly, our work thereby provides an empirical evidence that learning multiplicative interactions can result in a powerful inductive bias for increased robustness to sensor failures.


    Access

    Download


    Export, share and cite



    Title :

    Learning Multiplicative Interactions with Bayesian Neural Networks for Visual-Inertial Odometry


    Contributors:

    Conference:



    Publication date :

    2020-07-13


    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Radar aided visual inertial odometry initialization

    NIESEN URS | European Patent Office | 2020

    Free access

    RADAR AIDED VISUAL INERTIAL ODOMETRY INITIALIZATION

    NIESEN URS | European Patent Office | 2019

    Free access

    Attention Guided Unsupervised learning of Monocular Visual-inertial Odometry

    Wang, Zhenke / Zhu, Yuan / Lu, Ke et al. | IEEE | 2022


    RADAR AIDED VISUAL INERTIAL ODOMETRY OUTLIER REMOVAL

    NIESEN URS | European Patent Office | 2019

    Free access

    Control-enabled observability in visual-inertial odometry

    Bai, He / Taylor, Clark N. | IEEE | 2017