Point Cloud vs. Mesh Features for Building Interior Classification

Remote Sensing

Maarten Bassier¹  Maarten Vergauwen¹  Florent Poux² 

[1] Department of Civil Engineering, TC Construction—Geomatics, KU Leuven—Faculty of Engineering Technology, 9000 Ghent, Belgium;Geomatics Unit, University of Liège, 4000 Liège, Belgium;

关键词: feature extraction;    unsupervised segmentation;    classification;    machine learning;    BIM;    point clouds;   

DOI  :  10.3390/rs12142224

来源: DOAJ

【 摘 要 】

Interpreting 3D point cloud data of the interior and exterior of buildings is essential for automated navigation, interaction and 3D reconstruction. However, the direct exploitation of the geometry is challenging due to inherent obstacles such as noise, occlusions, sparsity or variance in the density. Alternatively, 3D mesh geometries derived from point clouds benefit from preprocessing routines that can surmount these obstacles and potentially result in more refined geometry and topology descriptions. In this article, we provide a rigorous comparison of both geometries for scene interpretation. We present an empirical study on the suitability of both geometries for the feature extraction and classification. More specifically, we study the impact for the retrieval of structural building components in a realistic environment which is a major endeavor in Building Information Modeling (BIM) reconstruction. The study runs on segment-based structuration of both geometries and shows that both achieve recognition rates over 75% F1 score when suitable features are used.

【 授权许可】

Unknown