Neural network pruning has become a de facto component for deploying deep networks on resource-constrained devices, which can reduce memory requirements and computation costs. In particular, channel pruning gained more popularity due to its structured nature and direct savings on general hardware. However, most existing pruning approaches utilize importance measures that are not directly related to the task utility. Moreover, few in the literature focus on visual detection models. To fill these gaps, we propose a novel gradient-based saliency measure for visual detection and use it to guide our channel pruning. Experiments on the KITTI and COCO_traffic datasets demonstrate our pruning method’s efficacy and superiority over competing state-of-the-art approaches. It can even achieve better performance with fewer parameters than the original model. Our pruning approach also demonstrates its great potential in handling small-scale objects.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Visual-Saliency-Guided Channel Pruning for Deep Visual Detectors in Autonomous Driving


    Contributors:
    Choi, Jung Im (author) / Tian, Qing (author)


    Publication date :

    2023-06-04


    Size :

    6268682 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    AUTONOMOUS VEHICLE OPERATIONAL MANAGEMENT WITH VISUAL SALIENCY PERCEPTION CONTROL

    NODA KUNIAKI / WRAY KYLE HOLLINS / WITWICKI STEFAN | European Patent Office | 2023

    Free access


    SaliencyVR: Saliency Matching Based Visual Relocalization for Autonomous Vehicle

    Wang, Ke / Zhao, Guoliang / Wang, Yong et al. | IEEE | 2025


    Autonomous vehicle operational management with visual saliency perception control

    NODA KUNIAKI / WRAY KYLE HOLLINS / WITWICKI STEFAN | European Patent Office | 2021

    Free access

    Visual Routines for Autonomous Driving

    Salgian, G. / Ballard, D. H. / IEEE; Computer Society | British Library Conference Proceedings | 1998