Representation learning in recent years has been addressed with self-supervised learning methods. The input data is augmented into two distorted views and an encoder learns the representations that are invariant to distortions – cross-view prediction. Augmentation is one of the key components in cross-view self-supervised learning frameworks to learn visual representations. This paper presents ExAgt, a novel method to include expert knowledge for augmenting traffic scenarios, to improve the learnt representations without any human annotation. The expert-guided augmentations are generated in an automated fashion based on the infrastructure, the interactions between the EGO and the traffic participants and an ideal sensor model. The ExAgt method is applied in two state-of-the-art cross-view prediction methods and the representations learnt are tested in downstream tasks like classification and clustering. Results show that the ExAgt method improves representation learning compared to using only standard augmentations and it provides a better representation space stability. The code is available at https://github.com/lab176344/ExAgt.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    ExAgt: Expert-guided Augmentation for Representation Learning of Traffic Scenarios


    Beteiligte:
    Balasubramanian, Lakshman (Autor:in) / Wurst, Jonas (Autor:in) / Egolf, Robin (Autor:in) / Botsch, Michael (Autor:in) / Utschick, Wolfgang (Autor:in) / Deng, Ke (Autor:in)


    Erscheinungsdatum :

    08.10.2022


    Format / Umfang :

    552651 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Expert-LaSTS: Expert-Knowledge Guided Latent Space for Traffic Scenarios

    Wurst, Jonas / Balasubramanian, Lakshman / Botsch, Michael et al. | IEEE | 2022


    Expert-LaSTS: Expert-Knowledge Guided Latent Space for Traffic Scenarios

    Wurst, Jonas / Balasubramanian, Lakshman / Botsch, Michael et al. | ArXiv | 2022

    Freier Zugriff

    Open-World Learning for Traffic Scenarios Categorisation

    Balasubramanian, Lakshman / Wurst, Jonas / Botsch, Michael et al. | IEEE | 2023



    LEARNING TO PREDICT LANE CHANGES IN HIGHWAY SCENARIOS USING DYNAMIC FILTERS ON A GENERIC TRAFFIC REPRESENTATION

    Mänttäri, Joonatan / Folkesson, John / Ward, Erik | British Library Conference Proceedings | 2018