3D Model Retrieval Based on Multi-Shell Extended Gaussian Image

In this paper, we consider a new shape representation for 3D object, called multi-resolution Multi-Shell Extended Gaussian Image (MSEGI) which eliminates the major drawback of EGI for not containing any direct distance information. MSEGI decomposes a 3D mesh model into multi-concentric shells by the normal distance of the outward surfaces to the origin, and captures the surface area distribution of a 3D model with surface orientation in each concentric shell. Then this distribution function is transformed to spherical harmonic coefficients which can provide multi-resolution shape descriptions by adopting different dimensions. Experimental results based on the public Princeton Shape Benchmark (PSB) dataset of 3D models show that the MSEGI significantly improves EGI, and outperforms CEGI and some popular shape descriptors.

[1]  Dietmar Saupe,et al.  Tools for 3D-object retrieval: Karhunen-Loeve transform and spherical harmonics , 2001, 2001 IEEE Fourth Workshop on Multimedia Signal Processing (Cat. No.01TH8564).

[2]  Berthold K. P. Horn.  Extended Gaussian images , 1984, Proceedings of the IEEE.

[3]  Hiroshi Matsuo,et al.  3-D object recognition using MEGI model from range data , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[4]  Bernard Chazelle,et al.  Matching 3D models with shape distributions , 2001, Proceedings International Conference on Shape Modeling and Applications.

[5]  Katsushi Ikeuchi,et al.  Determining 3-D object pose using the complex extended Gaussian image , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Ming Ouhyoung,et al.  On Visual Similarity Based 3D Model Retrieval , 2003, Comput. Graph. Forum.

[7]  Thomas A. Funkhouser,et al.  The Princeton Shape Benchmark , 2004, Proceedings Shape Modeling Applications, 2004..

[8]  Hau-San Wong,et al.  An evolutionary optimization approach for 3D human head model classification , 2003, MIR '03.

[9]  Szymon Rusinkiewicz,et al.  Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors , 2003, Symposium on Geometry Processing.

[10]  Philip E. Brou.  Using the Gaussian Image to Find the Orientation of Objects , 1984 .

[11]  Dejan V. VraniC.  An improvement of rotation invariant 3D-shape based on functions on concentric spheres , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[12]  Changming Sun,et al.  3D Symmetry Detection Using The Extended Gaussian Image , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

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