Segmentation from point cloud data is essential in many applications, such as remote sensing, mobile robots, or autonomous cars. However, the point clouds captured by the 3D range sensor are commonly sparse and unstructured, challenging efficient segmentation. A fast solution for point cloud instance segmentation with small computational demands is lacking. To this end, we propose a novel fast Euclidean clustering (FEC) algorithm which applies a point-wise scheme over the cluster-wise scheme used in existing works. The proposed method avoids traversing every point constantly in each nested loop, which is time and memory-consuming. Our approach is conceptually simple, easy to implement (40 lines in C++), and achieves two orders of magnitudes faster against the classical segmentation methods while producing high-quality results.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    FEC: Fast Euclidean Clustering for Point Cloud Segmentation


    Beteiligte:
    Yu Cao (Autor:in) / Yancheng Wang (Autor:in) / Yifei Xue (Autor:in) / Huiqing Zhang (Autor:in) / Yizhen Lao (Autor:in)


    Erscheinungsdatum :

    2022




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Unbekannt




    Fast-Spherical-Projection-Based Point Cloud Clustering Algorithm

    Chen, Zhihui / Xu, Hao / Zhao, Junxuan et al. | Transportation Research Record | 2022


    Fast Semantic Segmentation of 3D Lidar Point Cloud Based on Random Forest Method

    Jiang, Songdi / Guo, Wei / Fan, Yuzhi et al. | Springer Verlag | 2022


    Fast Semantic Segmentation of 3D Lidar Point Cloud Based on Random Forest Method

    Jiang, Songdi / Guo, Wei / Fan, Yuzhi et al. | TIBKAT | 2022


    A probability distribution-based point cloud clustering algorithm

    Yuan,X. / Zhao,C. / Zhang,H. et al. | Kraftfahrwesen | 2012


    A Fast Point Cloud Ground Segmentation Approach Based on Coarse-To-Fine Markov Random Field

    Huang, Weixin / Liang, Huawei / Lin, Linglong et al. | IEEE | 2022