We propose an enhancement to our previously proposed novel model called Contextual Vision Transformer (ViT) to address the problem of traffic accident risk forecasting. This framework combines spatial and temporal information using a data-driven approach. By treating the problem as a computer vision task, we can predict traffic accident risk as the next frame in a video sequence. Specificaly, we extend the ViT network with a Static Map generation (named XViT) for even better results on the Chicago dataset. Furthermore, we propose a Coarse-Fine-Coarse transformer architecture as an alternative approach to enhance traffic accident risk prediction.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Traffic Accident Risk Forecasting using Contextual Vision Transformers with Static Map Generation and Coarse-Fine-Coarse Transformers


    Beteiligte:


    Erscheinungsdatum :

    2023-09-24


    Format / Umfang :

    842068 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Traffic Accident Risk Forecasting using Contextual Vision Transformers

    Saleh, Khaled / Grigorev, Artur / Mihaita, Adriana-Simona | IEEE | 2022


    Flight Demand Forecasting with Transformers

    Wang, Liya / Mykityshyn, Amy / Johnson, Craig et al. | AIAA | 2022


    Coarse-to-fine stereo vision with accurate 3D boundaries

    Sizintsev, M. / Wildes, R. P. | British Library Online Contents | 2010


    A Coarse to Fine Matching Procedure for Linear Stereo Vision

    Burie, J.-C. / Postaire, J.-G. / IFAC | British Library Conference Proceedings | 1995