A boosting method based on SVM for relevance feedback in content-based 3D model retrieval

The technique of relevance feedback has been introduced to content-based 3D model retrieval. Support Vector Machine as a learner is one of the classical approaches in relevance feedback. And the Boosting method, as one of the ensemble methods, can establish a strong leaner by combing the component learners. In this paper, a novel relevance feedback mechanism, which makes use of the main idea of boosting and the component SVM, is presented and applied to the content-based 3D model retrieval. The experiments, based on the 3D model database Princeton Shape Benchmark, show that the relevance feedback algorithm can improve the retrieval performance of traditional SVM in 3D model retrieval.

[1]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[2]  Kagan Tumer,et al.  Error Correlation and Error Reduction in Ensemble Classifiers , 1996, Connect. Sci..

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

[4]  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).

[5]  Lars Kai Hansen,et al.  Neural Network Ensembles , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[7]  Bernhard Schölkopf,et al.  Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[8]  Zhengxin Chen,et al.  A Descriptive Framework for the Field of Data Mining and Knowledge Discovery , 2008, Int. J. Inf. Technol. Decis. Mak..

[9]  Yoav Freund,et al.  Boosting a weak learning algorithm by majority , 1990, COLT '90.

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

[11]  Raymond J. Mooney,et al.  Creating diversity in ensembles using artificial data , 2005, Inf. Fusion.

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

[13]  Indriyati Atmosukarto,et al.  Feature Combination and Relevance Feedback for 3D Model Retrieval , 2005, 11th International Multimedia Modelling Conference.

[14]  Tsuhan Chen,et al.  Feature space warping: an approach to relevance feedback , 2002, Proceedings. International Conference on Image Processing.

[15]  Manfred K. Warmuth,et al.  The weighted majority algorithm , 1989, 30th Annual Symposium on Foundations of Computer Science.

[16]  Ron Meir,et al.  Semantic-oriented 3d shape retrieval using relevance feedback , 2005, The Visual Computer.

[17]  Michael Elad,et al.  Content based retrieval of VRML objects: an iterative and interactive approach , 2002 .