Digital Curvatures Applied to 3D Object Analysis and Recognition: A Case Study

This paper uses digital curvatures for 3D object analysis and recognition. For direct adjacency in 3D, digital surface points have only six types. It is easy to determine and classify the digital curvatures of each point on the boundary of a 3D object. This is simpler than the case of triangulation on the boundary surface of a solid; the curvature can be any real value. This paper focuses on the global properties of categorizing curvatures for small regions. We use both digital Gaussian curvatures and digital mean curvatures to characterize 3D shapes. Then propose a multi-scale method and a feature vector method for 3D similarity measurement. We found that Gaussian curvatures mainly describe the global features and average characteristics such as the five regions of a human face. However, mean curvatures can be used to find local features and extreme points such as nose in 3D facial data.

[1]  Tamal K. Dey,et al.  Computing homology groups of simplicial complexes in R3 , 1998, JACM.

[2]  Ramesh C. Jain,et al.  Invariant surface characteristics for 3D object recognition in range images , 1985, Comput. Vis. Graph. Image Process..

[3]  Thomas A. Funkhouser,et al.  Shape-based retrieval and analysis of 3d models , 2005, CACM.

[4]  Miroslaw Bober,et al.  Curvature Scale Space Representation: Theory, Applications, and MPEG-7 Standardization , 2011, Computational Imaging and Vision.

[5]  Li Chen.  Genus Computing for 3D digital objects: algorithm and implementation , 2009, ArXiv.

[6]  Herbert Edelsbrunner,et al.  An incremental algorithm for Betti numbers of simplicial complexes on the 3-sphere , 1995, Comput. Aided Geom. Des..

[7]  Taku Komura,et al.  Topology matching for fully automatic similarity estimation of 3D shapes , 2001, SIGGRAPH.

[8]  Remco C. Veltkamp,et al.  A Survey of Content Based 3D Shape Retrieval Methods , 2004, SMI.

[9]  Guillaume Damiand,et al.  Computing Homology Generators for Volumes Using Minimal Generalized Maps , 2008, IWCIA.

[10]  Yongwu Rong,et al.  Digital topological method for computing genus and the Betti numbers , 2010 .

[11]  Reinhard Klette,et al.  Border and SurfaceTracing - Theoretical Foundations , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Mark Meyer,et al.  Discrete Differential-Geometry Operators for Triangulated 2-Manifolds , 2002, VisMath.

[13]  John M Sullivan,et al.  Cubic Polyhedra , 2002 .

[14]  Nasser Khalili,et al.  Estimation of Error in Curvature Computation on Multi-Scale Free-Form Surfaces , 2002, International Journal of Computer Vision.

[15]  Marcel Worring,et al.  Digital curvature estimation , 1993 .

[16]  Katsushi Ikeuchi,et al.  A Spherical Representation for Recognition of Free-Form Surfaces , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Rama Chellappa,et al.  Invariant Geometric Representation of 3D Point Clouds for Registration and Matching , 2006, 2006 International Conference on Image Processing.

[18]  John K. Tsotsos,et al.  Shape Representation and Recognition from Multiscale Curvature , 1997, Comput. Vis. Image Underst..

[19]  Ramesh C. Jain,et al.  Three-dimensional object recognition , 1985, CSUR.

[20]  Li Chen,et al.  Linear time recognition algorithms for topological invariants in 3D , 2008, 2008 19th International Conference on Pattern Recognition.

[21]  Jitendra Malik,et al.  Shape matching and object recognition using shape contexts , 2010, 2010 3rd International Conference on Computer Science and Information Technology.

[22]  Raimondo Schettini,et al.  3D face detection using curvature analysis , 2006, Pattern Recognit..

[23]  Azriel Rosenfeld,et al.  Digital geometry - geometric methods for digital picture analysis , 2004 .

[24]  Hiromi T. Tanaka,et al.  Curvature-based face surface recognition using spherical correlation-principal directions for curved object recognition , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[25]  Martial Hebert,et al.  On 3D shape similarity , 1996, Proceedings CVPR IEEE Computer Society Conference on Computer Vision and Pattern Recognition.