![](bg8.png)
5 CONCLUSION
In this paper a new unsupervised learning approach
for 3D point cloud segmentation is suggested which
uses a spectral clustering with higher eigenvectors in
combination with a decision tree. We described how
the new spectral clustering approach can be imple-
mented to obtain high quality results. When applying
this solution no a priori knowledge about the scene,
not even the number of clusters, is necessary and it is
shown that only a threshold for an objective function
has to be adapted in a few cases. Thus, the spectral
clustering method outperforms many other clustering
algorithms. This approach is very robust with respect
to various input data. Moreover in comparison to ot-
her methods the eigenvalues and eigenvectors are only
calculated once and then inserted into the segmenta-
tion tree. In the future we will use this clustering met-
hod as a pre-processing step for object detection and
pose estimation. Further adaption can be considered
regarding the similarity weight.
ACKNOWLEDGEMENTS
This work is supported by the European Social Fund
(ESF) and the Free State of Saxony.
REFERENCES
Arbelaez, P., Maire, M., Fowlkes, C., and Malik, J. (2011).
Contour detection and hierarchical image segmenta-
tion. IEEE Transactions on Pattern Analysis and Ma-
chine Intelligence, pages 898–916.
Cheng, J., Qiao, M., Bian, W., and Tao, D. (2011). 3D
human posture segmentation by spectral clustering
with surface normal constraint. Signal Processing,
91(9):2204–2212.
Comaniciu, D. and Meer, P. (2002). Mean shift: a ro-
bust approach toward feature space analysis. IEEE
Transactions on Pattern Analysis and Machine Intel-
ligence, 24(5):603–619.
Fiedler, M. (1975). A property of eigenvectors of non-
negative symmetric matrices and its application to
graph theory. Czechoslovak Mathematical Journal,
25(4):619–633.
Funk, E., Grießbach, D., Baumbach, D., Ernst, I., Boer-
ner, A., and Zuev, S. (2011). Segmentation of large
point-clouds using recursive local pca. International
Conference on Indoor Positioning and Navigation.
Hagen, L. and Kahng, A. B. (1992). New spectral methods
for ratio cut partitioning and clustering. IEEE Tran-
sactions on Computer-Aided Design of Integrated Ci-
rcuits and Systems, 11(9):1074–1085.
Hirschm
¨
uller, H. (2008). Stereo processing by semiglobal
matching and mutual information. IEEE Transacti-
ons on Pattern Analysis and Machine Intelligence,
30(2):328–341.
Hoiem, D., Efros, A. A., and Hebert, M. (2011). Recovering
occlusion boundaries from an image. International
Journal of Computer Vision, 91(3):328.
Karpathy, A., Miller, S., and Fei-Fei, L. (2013). Object
discovery in 3d scenes via shape analysis. In Interna-
tional Conference on Robotics and Automation.
Lai, Y.-K., Hu, S.-M., Martin, R. R., and Rosin, P. L. (2009).
Rapid and effective segmentation of 3D models using
random walks. Computer Aided Geometric Design,
26(6):665–679.
Liu, R. and Zhang, H. (2004). Segmentation of 3d mes-
hes through spectral clustering. In Proceedings of
the Computer Graphics and Applications, pages 298–
305. IEEE Computer Society.
Ma, T., Wu, Z., Feng, L., Luo, P., and Long, X. (2010).
Point cloud segmentation through spectral clustering.
In The 2nd International Conference on Information
Science and Engineering, pages 1–4.
Moumoun, L., Chahhou, M., Gadi, T., and Benslimane,
R. (2010). 3d hierarchical segmentation using the
markers for the watershed transformation. Internati-
onal Journal of Engineering Science and Technology,
2(7):3165–3171.
Papon, J., Abramov, A., Schoeler, M., and W
¨
org
¨
otter, F.
(2013). Voxel cloud connectivity segmentation - su-
pervoxels for point clouds. In Computer Vision and
Pattern Recognition (CVPR), 2013 IEEE Conference
on, pages 2027–2034.
Preetha, M. M. S. J., Suresh, L. P., and Bosco, M. J. (2012).
Image segmentation using seeded region growing. In
2012 International Conference on Computing, Elec-
tronics and Electrical Technologies, pages 576–583.
Richtsfeld, A., Morwald, T., Prankl, J., Zillich, M., and Vin-
cze, M. (2012). Segmentation of unknown objects in
indoor environments. In International Conference on
Intelligent Robots and Systems.
Rusu, R. B. (2009). Semantic 3D Object Maps for Everyday
Manipulation in Human Living Environments. PhD
thesis, Technische Universit
¨
at M
¨
unchen.
Rusu, R. B. and Cousins, S. (2011). 3D is here: Point Cloud
Library (PCL). In International Conference on Robo-
tics and Automation (ICRA), Shanghai, China.
Shi, J. and Malik, J. (2000). Normalized cuts and image
segmentation. IEEE Transactions on Pattern Analysis
and Machine Intelligence, 22(8):888–905.
Stein, S. C., Wrgtter, F., Schoeler, M., Papon, J., and Kul-
vicius, T. (2014). Convexity based object partitioning
for robot applications. IEEE International Conference
on Robotics and Automation, pages 3213–3220.
von Luxburg, U. (2007). A tutorial on spectral clustering.
Statistics and Computing, 17(4):395–416.
Wasenm
¨
uller, O. and Stricker, D. (2016). Comparison of
kinect v1 and v2 depth images in terms of accuracy
and precision. In ACCV Workshops.
Zelnik-Manor, L. and Perona, P. (2004). Self-tuning
spectral clustering. In Advances in Neural Informa-
tion Processing Systems 17, pages 1601–1608. MIT
Press.
VISAPP 2018 - International Conference on Computer Vision Theory and Applications
322