Comparison of surface normal estimation methods for range sensing applications

As mobile robotics is gradually moving towards a level of semantic environment understanding, robust 3D object recognition plays an increasingly important role. One of the most crucial prerequisites for object recognition is a set of fast algorithms for geometry segmentation and extraction, which in turn rely on surface normal vectors as a fundamental feature. Although there exists a plethora of different approaches for estimating normal vectors from 3D point clouds, it is largely unclear which methods are preferable for online processing on a mobile robot. This paper presents a detailed analysis and comparison of existing methods for surface normal estimation with a special emphasis on the trade-off between quality and speed. The study sheds light on the computational complexity as well as the qualitative differences between methods and provides guidelines on choosing the ‘right’ algorithm for the robotics practitioner. The robustness of the methods with respect to noise and neighborhood size is analyzed. All algorithms are benchmarked with simulated as well as real 3D laser data obtained from a mobile robot.

[1]  Tony DeRose,et al.  Surface reconstruction from unorganized points , 1992, SIGGRAPH.

[2]  Atsushi Nakazawa,et al.  Iterative refinement of range images with anisotropic error distribution , 2002 .

[3]  Chia-Hsiang Menq,et al.  Automatic data segmentation for geometric feature extraction from unorganized 3-D coordinate points , 2001, IEEE Trans. Robotics Autom..

[4]  M. Vanco.  A Direct Approach for the Segmentation of Unorganized Points and Recognition of Simple Algebraic Surfaces , 2002 .

[5]  Atsushi Nakazawa,et al.  Iterative refinement of range images with anisotropic error distribution , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[6]  Martin Buss,et al.  Realtime segmentation of range data using continuous nearest neighbors , 2009, 2009 IEEE International Conference on Robotics and Automation.

[7]  H. Gouraud.  Continuous Shading of Curved Surfaces , 1971, IEEE Transactions on Computers.

[8]  Hsi-Yung Feng,et al.  On the normal vector estimation for point cloud data from smooth surfaces , 2005, Comput. Aided Des..

[9]  Dirk Wollherr,et al.  A clustering method for efficient segmentation of 3D laser data , 2008, 2008 IEEE International Conference on Robotics and Automation.

[10]  Colin Bradley,et al.  Segmentation of a wrap-around model using an active contour , 1997, Comput. Aided Des..

[11]  Min Yang,et al.  Segmentation of measured point data using a parametric quadric surface approximation , 1999, Comput. Aided Des..

[12]  Tingting Xu,et al.  The autonomous city explorer project: aims and system overview , 2007, 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[13]  Anil K. Jain,et al.  Segmentation and Classification of Range Images , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Caihua Wang,et al.  Comparison of local plane fitting methods for range data , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[15]  Shuangshuang Jin,et al.  A comparison of algorithms for vertex normal computation , 2005, The Visual Computer.

[16]  Fisher.  Least Squares Optimization , 2004 .

[17]  Gene H. Golub,et al.  Matrix computations , 1983 .

[18]  Francis Sullivan,et al.  Least Squares Optimization , 1988 .

[19]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

[20]  金谷 健一.  Statistical optimization for geometric computation : theory and practice , 2005 .

[21]  Meenakshisundaram Gopi,et al.  Surface Reconstruction based on Lower Dimensional Localized Delaunay Triangulation , 2000, Comput. Graph. Forum.