Deep neural networks provide high accuracy for perception. However they require high computational power. In particular, LiDAR-based object detection delivers good accuracy and real-time performance, but demands high computation due to expensive feature-extraction from point cloud data in the encoder and backbone networks. We investigate the model complexity versus accuracy trade-off using reinforcement learning based pruning for PointPillars, a recent LiDAR-based 3D object detection network. We evaluate the model on the validation dataset of KITTI (80/20-splits) according to the mean average precision (mAP) for the car class. We prune the original PointPillars model (mAP 89.84) and achieve 65.8% reduction in floating point operations (FLOPs) for a marginal accuracy loss. The compression corresponds to 31.7% reduction in inference time and 35% reduction in GPU memory on GTX 1080 Ti.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Pruning CNNs for LiDAR-based Perception in Resource Constrained Environments




    Publication date :

    2021-07-11


    Size :

    3444533 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Dual-Branch CNNs for Vehicle Detection and Tracking on LiDAR Data

    Vaquero, Victor / del Pino, Ivan / Moreno-Noguer, Francesc et al. | IEEE | 2021


    LiDAR Imaging-based Attentive Perception

    Tsiourva, Maria / Papachristos, Christos | IEEE | 2020



    LiDAR-PDP: A LiDAR-Based Panoptic Dynamic Driving Environment Perception Algorithm

    Wang, Hai / Li, Jianguo / Cai, Yingfeng et al. | IEEE | 2025


    Perception system three lidar coverage

    IMPOLA TODD A / O'DONNELL TIMOTHY M / MARSOLEK JOHN L et al. | European Patent Office | 2023

    Free access