Abstract
Abstract. Point cloud data of indoor scenes is primarily composed of planar-dominant elements. Automatic shape segmentation is thus valuable to avoid labour intensive labelling. This paper provides a fully unsupervised region growing segmentation approach for efficient clustering of massive 3D point clouds. Our contribution targets a low-level grouping beneficial to object-based classification. We argue that the use of relevant segments for object-based classification has the potential to perform better in terms of recognition accuracy, computing time and lowers the manual labelling time needed. However, fully unsupervised approaches are rare due to a lack of proper generalisation of user-defined parameters. We propose a self-learning heuristic process to define optimal parameters, and we validate our method on a large and richly annotated dataset (S3DIS) yielding 88.1% average F1-score for object-based classification. It permits to automatically segment indoor point clouds with no prior knowledge at commercially viable performance and is the foundation for efficient indoor 3D modelling in cluttered point clouds.
Highlights
Automation in point cloud data processing is central for efficient decision-making systems and to cut labour costs
To quantify the quality of our segmentation approach, we first determine an "ideal" segmentation from the ground truth data by identifying connected components of points with the same classification. This segmentation is ideal in the sense that it contains the minimum number of segments while still allowing a per-segment classification with perfect accuracy
While strong oversegmentation will impact the performance of the subsequent classification step, it will usually not reduce the quality of the classification
Summary
Automation in point cloud data processing is central for efficient decision-making systems and to cut labour costs. Annotating each point by what it represents can be a long and tiresome job, to the point that the people doing it can unintentionally introduce errors through inattention or fatigue. It is cheaper and perhaps even more efficient to let a clustering algorithm group similar points together and only involve a human operator when assigning a label to the cluster. To this end, the underlying algorithm should provide relevant segments
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
