3D Object detection is a fundamental task in vision-based autonomous driving. Deep learning perception models achieve an outstanding performance at the expense of continuously increasing resource needs and, as such, increasing training costs. As inference time is still a priority, developers usually adopt a training pipeline where they first start using a compact architecture that yields a good trade-off between accuracy and latency. This architecture is usually found either by searching manually or by using neural architecture search approaches. Then, train the model and use light optimization techniques such as quantization to boost the model’s performance. In contrast, in this paper, we advocate for starting on a much larger model and then applying aggressive optimization to adapt the model to the resource-constraints. Our results on large-scale settings for 3D object detection demonstrate the benefits of initially focusing on maximizing the model’s accuracy and then achieving the latency requirements using network pruning.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Hardware-Aware Latency Pruning for Real-Time 3D Object Detection


    Beteiligte:
    Shen, Maying (Autor:in) / Mao, Lei (Autor:in) / Chen, Joshua (Autor:in) / Hsu, Justin (Autor:in) / Sun, Xinglong (Autor:in) / Knieps, Oliver (Autor:in) / Maxim, Carmen (Autor:in) / Alvarez, Jose M. (Autor:in)


    Erscheinungsdatum :

    04.06.2023


    Format / Umfang :

    19198314 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Real-time sleeper defect detection method based on pruning algorithm

    WU SONGRONG / TU ZHENWEI / YANG PING et al. | Europäisches Patentamt | 2022

    Freier Zugriff


    PRUNING SYSTEM USING FLYING OBJECT

    TERASAWA MASAAKI / AKITA MINORU / KOYANAGI RANSHI et al. | Europäisches Patentamt | 2024

    Freier Zugriff

    Rank-Based Filter Pruning for Real-Time UAV Tracking

    Wang, Xucheng / Zeng, Dan / Zhao, Qijun et al. | ArXiv | 2022

    Freier Zugriff

    OPTIMAL SENSOR READING FOR LOW LATENCY OBJECT DETECTION

    HART SEAN | Europäisches Patentamt | 2022

    Freier Zugriff