This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
The Probabilistic Peaking Effect of Viewed Angles and Distances with Application to 3-D Object Recognition
August 1990 (vol. 12 no. 8)
pp. 760-774

Two novel probabilistic models for viewed angles and distances are derived using an observability sphere method. The method, which is based on the assumption that the prior probability density is isotropic for all viewing orientations, can be used for the computation of observation probabilities for object's aspects, features, and probability densities of their quantitative attributes. Using the sphere, it is discovered that the probability densities of viewed angles, distances, and even projected curvature have sharp peaks at their original values. From this peaking effect, it is concluded that in most cases, the values of angles and distances are being altered only slightly by the imaging process, and they can still serve as a strong cue for model-based recognition. The probabilistic models for 3-D object recognition from monocular images are used. To form the angular elements that are needed, the objects are represented by their linear features and specific points primitives. Using the joint density model of angles and distances, the probabilities of initial matching hypotheses and mutual information coefficients are estimated. These results are then used for object recognition by optimal matching search and stochastic labeling schemes. Various synthetic and real objects are recognized by this approach.

[1] J. Ben-Arie, "The properties in viewed angles and distances with application to 3-D object recognition," inProc. ICPR Conf., Rome, Italy, Nov. 1988, pp. 309-312.
[2] J. Ben-Arie, "A vision-system: 3-D object recognition from 2-D single images" (in Hebrew), Dep. Aeronaut. Eng., Technion-I.I.T., Haifa, Res. Rep., Mar. 1988.
[3] J. Ben-Arie, "Computer vision: 3-D object recognition from 2-D monocular images, Part II" (in Hebrew), Dep. Aeronaut. Eng., Technion-I.I.T., Haifa, Res. Rep., July 1986.
[4] A. P. Witkin, "Recovering surface shape and orientation from texture,"Artificial Intell., vol. 17, pp. 17-47, 1981.
[5] T. Kanade, "Recovery of the three-dimensional shape of an object from a single view,"Artificial Intell., vol. 17, pp. 409-460, 1981.
[6] K. A. Stevens, "Representing and analyzing surface orientation," inArtificial Intelligence: An MIT Perspective, Vol. 2, P. H. Winston and R. H. Brown, Eds. Cambridge, MA: M.I.T. Press, 1979.
[7] M. F. Augusteijn and C. R. Dyer, "Model based shape from contour and point patterns," inProc. IEEE Conf. Comput. Vision Pattern Recognition, San Francisco, CA, June 1985, pp. 100-105.
[8] O. D. Faugeras and M. Berthod, "Improving consistency and reducing ambiguity in stochastic labeling: An optimization approach,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-3, July 1981.
[9] J. Ben-Arie and A. Z. Meiri, "3-D object recognition by optimal matching search of multinary relations-graphs,"Comput. Vision, Graphics, Image Processing, vol. 37, pp. 345-361, Mar. 1987.
[10] R. M. Haralick and L. G. Shapiro, "The consistent labeling problem--Part II,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-1, pp. 173-184, Apr. 1979.
[11] A. Rosenfeld, R. A. Hummel, and S. W. Zucker, "Scene labeling by relaxation operations,"IEEE Trans. Syst., Man, Cybern., vol. SMC-6, pp. 420-433, 1976.
[12] S. Peleg, "A new probabilistic relaxation scheme,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-2, July 1980.
[13] D. Hottenlocher and S. Ullman, "Object recognition using alignment," inProc. DARPA Image Understanding Workshop, Los Angeles, CA, Feb. 1987.
[14] J. Ben-Arie, "Probabilistic models of observed features and aspects with application to weighted aspect graphs,"Patt. Recogn. Lett., vol. 11, pp. 421-427, 1990.

Index Terms:
pattern recognition; probabilistic peaking effect; viewed angles; 3-D object recognition; observability sphere method; observation probabilities; probability densities; quantitative attributes; model-based recognition; linear features; specific points primitives; initial matching hypotheses; mutual information coefficients; optimal matching search; stochastic labeling schemes; pattern recognition; probability
Citation:
J. Ben-Arie, "The Probabilistic Peaking Effect of Viewed Angles and Distances with Application to 3-D Object Recognition," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, no. 8, pp. 760-774, Aug. 1990, doi:10.1109/34.57667
Usage of this product signifies your acceptance of the Terms of Use.