Convolutional Neural Networks (CNNs) are successfully used for the important automotive visual perception tasks including object recognition, motion and depth estimation, visual SLAM, etc. However, these tasks are typically independently explored and modeled. In this paper, we propose a joint multi-task network design for learning several tasks simultaneously. Our main motivation is the computational efficiency achieved by sharing the expensive initial convolutional layers between all tasks. Indeed, the main bottleneck in automated driving systems is the limited processing power available on deployment hardware. There is also some evidence for other benefits in improving accuracy for some tasks and easing development effort. It also offers scalability to add more tasks leveraging existing features and achieving better generalization. We survey various CNN based solutions for visual perception tasks in automated driving. Then we propose a unified CNN model for the important tasks and discuss several advanced optimization and architecture design techniques to improve the baseline model. The paper is partly review and partly positional with demonstration of several preliminary results promising for future research. We first demonstrate results of multi-stream learning and auxiliary learning which are important ingredients to scale to a large multi-task model. Finally, we implement a two-stream three-task network which performs better in many cases compared to their corresponding single-task models, while maintaining network size.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    NeurAll: Towards a Unified Visual Perception Model for Automated Driving


    Beteiligte:
    Sistu, Ganesh (Autor:in) / Leang, Isabelle (Autor:in) / Chennupati, Sumanth (Autor:in) / Yogamani, Senthil (Autor:in) / Hughes, Ciaran (Autor:in) / Milz, Stefan (Autor:in) / Rawashdeh, Samir (Autor:in)


    Erscheinungsdatum :

    01.10.2019


    Format / Umfang :

    901423 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    TOWARDS ROBUST DIRECT PERCEPTION NETWORKS FOR AUTOMATED DRIVING

    Cheng, Chih-Hong | British Library Conference Proceedings | 2020


    Systematization of Corner Cases for Visual Perception in Automated Driving

    Breitenstein, Jasmin / Termohlen, Jan-Aike / Lipinski, Daniel et al. | IEEE | 2020


    SYSTEMATIZATION OF CORNER CASES FOR VISUAL PERCEPTION IN AUTOMATED DRIVING

    Breitenstein, Jasmin / Termöhlen, Jan-Aike / Lipinski, Daniel et al. | British Library Conference Proceedings | 2020