Learning View Generalization Functions

Learning object models from views in 3D visual object recognition is usually formulated either as a function approximation problem of a function describing the view-manifold of an object, or as that of learning a class-conditional density. This paper describes an alternative framework for learning in visual object recognition, that of learning the view-generalization function. Using the view-generalization function, an observer can perform Bayes-optimal 3D object recognition given one or more 2D training views directly, without the need for a separate model acquisition step. The paper shows that view generalization functions can be computationally practical by restating two widely-used methods, the eigenspace and linear combination of views approaches, in a view generalization framework. The paper relates the approach to recent methods for object recognition based on non-uniform blurring. The paper presents results both on simulated 3D ``paperclip'' objects and real-world images from the COIL-100 database showing that useful view-generalization functions can be realistically be learned from a comparatively small number of training examples.

[1]  Ronen Basri,et al.  Recognition by Linear Combinations of Models , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  T. Poggio,et al.  A network that learns to recognize three-dimensional objects , 1990, Nature.

[3]  Massimiliano Pontil,et al.  Support Vector Machines for 3D Object Recognition , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  H. C. Longuet-Higgins,et al.  A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[5]  David C. Knill,et al.  Object classification for human and ideal observers , 1995, Vision Research.

[6]  Tomaso A. Poggio,et al.  Linear Object Classes and Image Synthesis From a Single Example Image , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  David G. Stork,et al.  Pattern Classification (2nd ed.) , 1999 .

[8]  Jitendra Malik,et al.  Geometric blur for template matching , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[9]  S. Edelman,et al.  Computational Theories of Object Recognition Edelman -computation and Object Recognition Ii Box 1. Structural Descriptions ~ 7~ Recognition by Components Varieties of Alignment Multidimensional Histograms Approximation in Feature Space , 2022 .

[10]  Shigeo Abe DrEng Pattern Classification , 2001, Springer London.

[11]  Thomas M. Breuel Character recognition by adaptive statistical similarity , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[12]  David G. Stork,et al.  Pattern Classification , 1973 .

[13]  Hiroshi Murase,et al.  Real-time 100 object recognition system , 1996, Proceedings of IEEE International Conference on Robotics and Automation.

[14]  Alex Pentland,et al.  View-based and modular eigenspaces for face recognition , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[15]  P. McCullagh Tensor Methods in Statistics , 1987 .

[16]  Sameer A. Nene,et al.  Columbia Object Image Library (COIL100) , 1996 .

[17]  David G. Lowe,et al.  Similarity Metric Learning for a Variable-Kernel Classifier , 1995, Neural Computation.