Neural networks are currently suggested to be implemented in several different driving functions of autonomous vehicles. While showing promising results the drawback lies in the difficulty of safety verification and ensuring operation as intended. The aim of this paper is to increase safety when using neural networks, by proposing a monitoring framework based on novelty estimation of incoming driving data. The idea is to use unsupervised instance discrimination to learn a similarity measure across ego-vehicle camera images. By estimating a von Mises-Fisher distribution of expected ego-camera images they can be compared with unexpected novel images. A novelty measurement is inferred through the likelihood of test frames belonging to the expected distribution. The suggested method provides competitive results to several other novelty or anomaly detection algorithms on the CIFAR-10 and CIFAR-100 datasets. It also shows promising results on real world driving scenarios by distinguishing novel driving scenes from the training data of BDD100 k. Applied on the identical training-test data split, the method is also able to predict the performance profile of a segmentation network. Finally, examples are provided on how this method can be extended to find novel segments in images.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Safety Monitoring of Neural Networks Using Unsupervised Feature Learning and Novelty Estimation


    Beteiligte:
    Ranjbar, Arian (Autor:in) / Hornauer, Sascha (Autor:in) / Fredriksson, Jonas (Autor:in) / Yu, Stella X. (Autor:in) / Chan, Ching-Yao (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2022-09-01


    Format / Umfang :

    6911772 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Scene Novelty Prediction from Unsupervised Discriminative Feature Learning

    Ranjbar, Arian / Yeh, Chun-Hsiao / Hornauer, Sascha et al. | IEEE | 2020


    Novelty Detection with Deep Learning

    Martinez, Jose / Donati, Alessandro | AIAA | 2018



    Safety Expertise and the Perils of Novelty

    McCormick, Frank / Graydon, Mallory / Neogi, Natasha et al. | IEEE | 2023


    Unsupervised Feature Learning for Dense Correspondences Across Scenes

    Zhang, C. / Shen, C. / Shen, T. | British Library Online Contents | 2016