Robust Classification of Curvilinear and Surface-Like Structures in 3d Point Cloud Data
The classification of 3d point cloud data is an important component of applications such as map generation and architectural modeling. However, the complexity of the scenes together with the level of noise in the data acquired through mobile laser range-scanning make this task quite difficult. We propose a novel classification method that relies on a combination of edge, node, and relative density information within an Associative Markov Network framework. The main application of our work is the classification of the structures within a point cloud into curvilinear, surface-like, and noise components. We are able to robustly extract complicated structures such as tree branches. The measures taken to ensure the robustness of our method generalize and can be leveraged in noise reduction applications as well. We compare our work with another state of the art classification technique, namely Directional Associative Markov Network, and show that our method can achieve significantly higher accuracy in the classification of the 3d point clouds.
KeywordsPoint Cloud Connected Structure Point Cloud Data Robust Principal Component Analysis LiDAR Sensor
Unable to display preview. Download preview PDF.
- 1.Unnikrishnan, R., Lalonde, J.F., Vandapel, N., Hebert, M.: Scale selection for the analysis of point-sampled curves. In: Proceedings of the Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT 2006), pp. 1026–1033. IEEE Computer Society, Washington, DC, USA (2006)CrossRefGoogle Scholar
- 2.Golovinskiy, A., Kim, V.G., Funkhouser, T.: Shape-based recognition of 3D point clouds in urban environments. In: International Conference on Computer Vision, ICCV (2009)Google Scholar
- 3.Munoz, D., Vandapel, N., Hebert, M.: Directional associative markov network for 3-d point cloud classification. In: Fourth International Symposium on 3D Data Processing, Visualization and Transmission (2008)Google Scholar
- 5.Munoz, D., Vandapel, N., Hebert, M.: Onboard contextual classification of 3-d point clouds with learned high-order markov random fields. In: Proceedings of the 2009 IEEE international conference on Robotics and Automation, ICRA 2009, pp. 4273–4280. IEEE Press, Piscataway (2009)Google Scholar
- 7.Lalonde, J.F., Unnikrishnan, R., Vandapel, N., Hebert, M.: Scale selection for classification of point-sampled 3-d surfaces. In: Proceedings of the Fifth International Conference on 3-D Digital Imaging and Modeling, pp. 285–292. IEEE Computer Society, Washington, DC, USA (2005)CrossRefGoogle Scholar
- 8.Xiong, X., Munoz, D., Bagnell, J.A.D., Hebert, M.: 3-d scene analysis via sequenced predictions over points and regions. In: IEEE International Conference on Robotics and Automation, ICRA (2011)Google Scholar
- 9.Munoz, D., Bagnell, J.A.D., Vandapel, N., Hebert, M.: Contextual classification with functional max-margin markov networks. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR (2009)Google Scholar
- 10.Shapovalov, R., Velizhev, A., Barinova, O.: Non-associative markov networks for 3d point cloud classification. Networks XXXVIII, 103–108 (2010)Google Scholar
- 11.Boykov, Y., Veksler, O., Zabih, R.: Efficient approximate energy minimization via graph cuts. IEEE Transactions on Pattern Analysis and Machine Intelligence (2001)Google Scholar
- 12.Taskar, B., Chatalbashev, V., Koller, D.: Learning associative markov networks. In: International Conference on Machine LearningGoogle Scholar
- 14.Candès, E.J., Li, X., Ma, Y., Wright, J.: Robust principal component analysis? J. ACM 58, 11:1–11:37 (2011)Google Scholar