ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5/W3, 191-197, 2015
http://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/II-5-W3/191/2015/
doi:10.5194/isprsannals-II-5-W3-191-2015
© Author(s) 2015. This work is distributed
under the Creative Commons Attribution 3.0 License.
 
11 Aug 2015
Point clouds segmentation as base for as-built BIM creation
H. Macher, T. Landes, and P. Grussenmeyer Nat. Inst. of Applied Sciences of Strasbourg, Photogrammetry & Geomatics Group, Strasbourg, France
Keywords: 3D modelling, indoor point clouds, space segmentation, as-built BIM, existing buildings, automation Abstract. In this paper, a three steps segmentation approach is proposed in order to create 3D models from point clouds acquired by TLS inside buildings. The three scales of segmentation are floors, rooms and planes composing the rooms. First, floor segmentation is performed based on analysis of point distribution along Z axis. Then, for each floor, room segmentation is achieved considering a slice of point cloud at ceiling level. Finally, planes are segmented for each room, and planes corresponding to ceilings and floors are identified. Results of each step are analysed and potential improvements are proposed. Based on segmented point clouds, the creation of as-built BIM is considered in a future work section. Not only the classification of planes into several categories is proposed, but the potential use of point clouds acquired outside buildings is also considered.
Conference paper (PDF, 1189 KB)


Citation: Macher, H., Landes, T., and Grussenmeyer, P.: Point clouds segmentation as base for as-built BIM creation, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5/W3, 191-197, doi:10.5194/isprsannals-II-5-W3-191-2015, 2015.

BibTeX EndNote Reference Manager XML