TY - GEN
T1 - Noise-resistant unsupervised object segmentation in multi-view indoor point clouds
AU - Bobkov, Dmytro
AU - Chen, Sili
AU - Kiechle, Martin
AU - Hilsenbeck, Sebastian
AU - Steinbach, Eckehard
N1 - Publisher Copyright:
Copyright © 2017 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.
PY - 2017
Y1 - 2017
N2 - 3D object segmentation in indoor multi-view point clouds (MVPC) is challenged by a high noise level, varying point density and registration artifacts. This severely deteriorates the segmentation performance of state-of-the-art algorithms in concave and highly-curved point set neighborhoods, because concave regions normally serve as evidence for object boundaries. To address this issue, we derive a novel robust criterion to detect and remove such regions prior to segmentation so that noise modelling is not required anymore. Thus, a significant number of inter-object connections can be removed and the graph partitioning problem becomes simpler. After initial segmentation, such regions are labelled using a novel recovery procedure. Our approach has been experimentally validated within a typical segmentation pipeline on multi-view and single-view point cloud data. To foster further research, we make the labelled MVPC dataset public (Bobkov et al., 2017).
AB - 3D object segmentation in indoor multi-view point clouds (MVPC) is challenged by a high noise level, varying point density and registration artifacts. This severely deteriorates the segmentation performance of state-of-the-art algorithms in concave and highly-curved point set neighborhoods, because concave regions normally serve as evidence for object boundaries. To address this issue, we derive a novel robust criterion to detect and remove such regions prior to segmentation so that noise modelling is not required anymore. Thus, a significant number of inter-object connections can be removed and the graph partitioning problem becomes simpler. After initial segmentation, such regions are labelled using a novel recovery procedure. Our approach has been experimentally validated within a typical segmentation pipeline on multi-view and single-view point cloud data. To foster further research, we make the labelled MVPC dataset public (Bobkov et al., 2017).
KW - Concavity criterion
KW - Laser scanner
KW - Object segmentation
KW - Point cloud
KW - Segmentation dataset
UR - http://www.scopus.com/inward/record.url?scp=85030219668&partnerID=8YFLogxK
U2 - 10.5220/0006100801490156
DO - 10.5220/0006100801490156
M3 - Conference contribution
AN - SCOPUS:85030219668
T3 - VISIGRAPP 2017 - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
SP - 149
EP - 156
BT - VISAPP
A2 - Imai, Francisco
A2 - Tremeau, Alain
A2 - Braz, Jose
PB - SciTePress
T2 - 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2017
Y2 - 27 February 2017 through 1 March 2017
ER -