Semantic Interpretation of Mobile Laser Scanner Point Clouds in Indoor Scenes Using Trajectories

Abstract

The data acquisition with Indoor Mobile Laser Scanners (IMLS) is quick, low-cost and accurate for indoor 3D modeling. Besides a point cloud, an IMLS also provides the trajectory of the mobile scanner. We analyze this trajectory jointly with the point cloud to support the labeling of noisy, highly reflected and cluttered points in indoor scenes. An adjacency-graph-based method is presented for detecting and labeling of permanent structures, such as walls, floors, ceilings, and stairs. Through occlusion reasoning and the use of the trajectory as a set of scanner positions, gaps are discriminated from real openings in the data. Furthermore, a voxel-based method is applied for labeling of navigable space and separating them from obstacles. The results show that 80% of the doors and 85% of the rooms are correctly detected, and most of the walls and openings are reconstructed. The experimental outcomes indicate that the trajectory of MLS systems plays an essential role in the understanding of indoor scenes. the complexity of 3D space. Few works deal with arbitrary wall layouts [6][7][8], but they are restricted to vertical walls and horizontal ceilings. Our method detects slanted walls and sloped ceilings exploiting the adjacency of permanent structures, based on the assumption that there is less clutter near the ceiling in indoor environments. Additionally, the arbitrary arrangements of walls (non-Manhattan-World) will be handled in this work. Our pipeline for semantic labeling of permanent structure uses detection of planar primitives labelled as wall, floor and ceiling, and their topological relations.Room segmentation is another research problem in large-scale indoor modeling. In the literature, different approaches, such as Voronoi graphs, cell decomposition, binary space partitioning and morphology operators [9] are suggested for 2D and 3D room segmentation. Some of these methods have limitations, such as Manhattan-World constraints and vertical walls. Most of the room segmentation methods rely on the viewpoint [8,10] and require scanning with a TLS in each room [7]. However, as opposed to one scanning location per room, mobile laser scanning systems produce a continuous trajectory and assigning points per room based on the scan location is not possible. Similar to our method for trajectory analysis, refs. [11,12] exploit the trajectory for space subdivision. Although their focus is only on space subdivision and simple structure, their results support our motivation of using the trajectory for interpretation of point clouds.In our pipeline, a novel method is suggested for partitioning interior spaces based on voxels and exploiting unoccupied space. Besides knowing the room layout, information about the doors, walkable space and stairs supports navigation planning. Therefore, voxels are used to identify the walkable space and the trajectory to identify the stairs and doors.Reflective surfaces, such as glass, complicate the analysis of indoor point clouds. Such surfaces cause the appearance of "ghost walls" in the...

Indoor 3d Modeling and Flexible Space Subdivision From Point Clouds

“…Other works reduce the complexity of the structure assuming walls are always vertical or ceilings are at the same height which generates a 2.5D model (Ikehata et al, 2015;Tran H. et al, 2019;Turner et al, 2015). Few works deal with arbitrary wall layout and slanted walls (Li et al, 2018;Mura et al, 2016;Nikoohemat et al, 2018). The process of 3D modelling from the point clouds is not limited to the detection of the permanent structures but also reconstruction of a watertight model.…”

“…Machine learning and deep learning algorithms are used in some works for detection of indoor objects such as walls, windows and doors (Boulch et al, 2018;Qi et al, 2016). In this work, the adjacency graph developed by (Nikoohemat et al, 2018) is used for detection of the permanent structure and the trajectory of the mobile laser scanners is exploited for identifying the doors.…”

“…The agent is the actor who should be guided during the disaster. Recent literature addresses the problem of indoor 3D reconstruction from point clouds in a different pipeline than navigation, mainly for BIM applications (Ikehata et al, 2015;Mura et al, 2016;Nikoohemat et al, 2018;Ochmann et al, 2016). On the other hand, there are some related work focusing on the problem of path finding and navigation inside the buildings using already existing models (Diakité and Zlatanova, 2018;Nagel, 2014;Wang and Zipf, 2017).…”

“…In chapter 3, the problem of modelling from point clouds is tackled. The represented method is built on the pipeline suggested by (Nikoohemat et al, 2018). Chapter 4 explains how to generate a Flexible Space Subdivision (FSS) (Diakité and Zlatanova, 2018) from the 3D model generated in chapter 3.…”

Obstacle-Aware Indoor Pathfinding Using Point Clouds

“…Trajectory was also used by Reference [27] to label doors, floors, walls, and ceilings in indoor spaces. This work was extended in Reference [28], in which the indoor space was subdivided by implementing morphological operations and connected components.…”