During the ISPRS Geospatial Week 2017 which took place in Wuhan, China, PhD candidate Shayan Nikoohemat received the "Best youth oral paper award" as well as the overall "Best paper award".
The paper titled "Exploiting Indoor Mobile Laser Scanner Trajectory for Semantic Interpretation of Point Clouds" was selected for the "Best youth oral paper award" by the chairs of the Indoor3D workshop based on the results of the peer review process. The co-authors next to Shayan are Dr. Michael Peter, Dr. Sander Oude Elberink, and Prof. George Vosselman. Among the awardees of all workshops of the Geospatial Week 2017 the paper was additionally awarded as "Best paper" by the organizing committee of the event.
The paper describes several algorithms for the extraction of semantic information from point clouds captured in indoor environments using mobile mapping systems. In addition to the point clouds it especially concentrates on the use of the system's trajectory as a valuable information source. Methods for the robust labelling of points as belonging to walls, floors, or ceilings, but also the detection of openings, i.e. doors and windows, and the labelling of space as navigable are presented. The research was carried out in the context of the NWO-TTW Maps4Society project SIMs3D in which methods for the reconstruction of as-built indoor models for use in emergency cases are developed.
Link to the full paper: https://doi.org/10.5194/isprs-annals-IV-2-W4-355-2017
Figure 1: a) Input data disturbed by clutter (e.g. furniture), occlusions, reflections; b) detected walls (orange), openings (cyan), incorrect walls (dotted black) and openings (solid black); result after automatic cleaning
