This paper presents a novel CNN-based approach for synthesizing high-resolution LiDAR point cloud data. Our approach generates semantically and perceptually realistic results with guidance from specialized loss-functions. First, we utilize a modified per-point loss that addresses missing LiDAR point measurements. Second, we align the quality of our generated output with real-world sensor data by applying a perceptual loss. In large-scale experiments on real-world datasets, we evaluate both the geometric accuracy and semantic segmentation performance using our generated data vs. ground truth. In a mean opinion score testing we further assess the perceptual quality of our generated point clouds. Our results demonstrate a significant quantitative and qualitative improvement in both geometry and semantics over traditional non CNN-based upsampling methods.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    CNN-based synthesis of realistic high-resolution LiDAR data


    Contributors:


    Publication date :

    2019-06-01


    Size :

    1350249 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    MARSIM: A light-weight point-realistic simulator for LiDAR-based UAVs

    Kong, Fanze / Liu, Xiyuan / Tang, Benxu et al. | ArXiv | 2022

    Free access

    Synthetic vision helicopter flights using high resolution LIDAR terrain data

    Sindlinger, A. / Meuter, M. / Barraci, N. et al. | SPIE | 2006


    Autoregistration of high-resolution satellite imagery using LIDAR intensity data

    Lee, Jaebin / Lee, Changno / Yu, Kiyun | Springer Verlag | 2011


    High Resolution LiDAR Based on Single Chip SPAD Array

    Yanai, Kenichi / Azuma, Kenta / Ozaki, Noriyuki et al. | SAE Technical Papers | 2019