This Article 
 Bibliographic References 
 Add to: 
Adaptive 3-D Object Recognition from Multiple Views
February 1992 (vol. 14 no. 2)
pp. 107-124

The authors address the problem of generating representations of 3-D objects automatically from exploratory view sequences of unoccluded objects. In building the models, processed frames of a video sequence are clustered into view categories called aspects, which represent characteristic views of an object invariant to its apparent position, size, 2-D orientation, and limited foreshortening deformation. The aspects as well as the aspect transitions of a view sequence are used to build (and refine) the 3-D object representations online in the form of aspect-transition matrices. Recognition emerges as the hypothesis that has accumulated the maximum evidence at each moment. The 'winning' object continues to refine its representation until either the camera is redirected or another hypothesis accumulates greater evidence. This work concentrates on 3-D appearance modeling and succeeds under favorable viewing conditions by using simplified processes to segment objects from the scene and derive the spatial agreement of object features.

[1] A. A. Baloch and A. M. Waxman, "Visual learning, adaptive expectations. and behavioral conditioning of the mobile robot MAVIN,"Neural Networks, vol. 4, no. 3, pp. 271-302, 1991.
[2] K. Bowyer, D. Eggert, J. Stewman, and L. Stark, "Developing the aspect graph representation for use in image understanding," inProc. 1989 Image Understanding Workshop, 1989, pp. 831-849.
[3] G. A. Carpenter and S. Grossberg, "ART 2: Self-organization of stable category recognition codes for analog input patterns,"Appl. Opt., vol. 26, no. 23, pp. 4919-4930, 1987.
[4] S. Chen and H. Freeman, "Computing characteristic views of quadric-surfaced solids," inProc. 10th Int. Conf. Patt. Recogn., 1990.
[5] S. Edelman, H. Bülthoff, and D. Weinshall, "Stimulus familiarity determines recognition strategy for novel 3D objects," MIT AI Lab Memo no. 1138, Cambridge, MA, 1989.
[6] G. Fekete and L. S. Davis, "Property spheres: A new representation for 3-D object representation," inProc. 1984 IEEE Workshop Comput. Vision: Representation Contr., 1984, pp. 192-201.
[7] H. Freeman and I. Chakravarty, "The use of characteristic views in the recognition of three-dimensional objects," inPattern Recognition in Practice(E. S. Gelsema and L. N. Kanal, Eds.). New York: North Holland, 1980.
[8] Z. Gigus and J. Malik, "Computing the aspect graph for line drawings of polyhedral objects," inProc. IEEE Comp. Soc. Conf. Computer Vision and Pattern Recognition, IEEE, New York, June 1988, pp. 654-661.
[9] S. Grossberg, "Contour enhancement, short term memory, and constancies in reverberating neural networks,"Studies Appl. Math., vol. 52, no. 3, pp. 217-257, 1973.
[10] L. Kitchen and A. Rosenfeld, "Grey-level corner detection,"Patt. Recogn. Lett., vol. 1, no. 2, pp. 95-102, 1982.
[11] J. J. Koenderink and A. J. van Doorn, "The internal representation of solid shape with respect to vision,"Biolog. Cybern., vol. 32, pp. 211-216, 1979.
[12] J. Lazzaro, S. Ryckebusch, M. A. Mahowald, and C. A. Mead, "Winnertake-all networks ofO(N)complexity," inAdvances in Neural Information Processing Systems 1(D. S. Touretzky, Ed.). San Mateo, CA: Morgan Kaufmann, 1989, pp. 703-711.
[13] C. -H. Liu and W. -H. Tsai, "3D curved object recognition from multiple 2D camera views,"Comput. Vision Graphics Image Processing, vol. 50, pp. 177-187, 1990.
[14] K. V. Mardia,Statistics of Directional Data. New York: Academic, 1972.
[15] W. N. Martin and J. K. Aggarwal, "Volumetric descriptions of objects from multiple views,"IEEE Trans. Patt. Anal. Machine Intell., vol. 5, no. 2, pp. 150-158, 1983.
[16] D. I. Perrettet al., "Frameworks of analysis for the neural representations of animate objects and actions,"J. Exper. Biology, vol. 146, pp. 87-113, 1989.
[17] D. I. Perrett, A. J. Mistlin, and A. J. Chitty, "Visual neurones responsive to faces,"Trends Neuroscience, vol. 10, no. 9, pp. 358-363, 1987.
[18] D. I. Perrettet al., "Visual cells in the temporal cortex sensitive to face view and gaze direction," inProc. R. Soc. London, Series B, vol. 223, pp. 293-317, 1985.
[19] W.H. Plantinga and C.R. Dyer, "Visibility, occlusion and the aspect graph,"Int. J. Comput. Vision, 1990.
[20] J. Ponce and D. J. Kriegman, "Computing exact aspect graphs of curved objects: Parametric surfaces," inProc. 8th Nat. Conf. Artificial Intell., 1990, pp. 1074-1079.
[21] J. H. Rieger, "The geometry of view space opaque objects bounded by smooth surfaces,"Artificial Intell., vol. 44, pp. 1-40, 1990.
[22] R. D. Rimey and C. M. Brown, "HMM's and vision: Representing structure and sequences for active vision using hidden markov models," Comput. Sci. Dept., Univ. of Rochester, TR 366, Jan. 1991.
[23] A. Rosenfeld, "Recognizing unexpected objects: A proposed approach,"Int. J. Pattern Recognition Artificial Intell., vol. 1, pp. 71-84, 1987.
[24] M. Seibert, A. A. Baloch, and A. M. Waxman,Modular Neural Systems: Visual Learning of 3D Objects and Conditioning of a Mobile Robot.Elmsford, NY: Pergammon, 1992.
[25] M. Seibert and A. M. Waxman, "Spreading activation layers, visual saccades, and invariant representations for neural pattern recognition systems,"Neural Networks, vol. 2, no. 1, pp. 9-27, 1989.
[26] T. Sripradisvarakul and R. Jain, "Generating aspect graphs for curved objects," inProc. IEEE Workshop Interpretation of 3D Scenes, Nov. 1989, pp. 109-115.
[27] S. A. Underwood and C. L. Coates, "Visual learning from multiple views,"IEEE Trans. Comput., vol. C-24, no. 6, pp. 651-661, 1975.
[28] P. H. Winston,The Psychology of Computer Vision. New York: McGraw-Hill, 1975.

Index Terms:
3D object adaptive recognition; pattern recognition; clustering; 3D appearance modelling; segmentation; exploratory view sequences; aspect-transition matrices; adaptive systems; pattern recognition; picture processing
M. Seibert, A.M. Waxman, "Adaptive 3-D Object Recognition from Multiple Views," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 107-124, Feb. 1992, doi:10.1109/34.121784
Usage of this product signifies your acceptance of the Terms of Use.