This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Context-Based Vision: Recognizing Objects Using Information from Both 2D and 3D Imagery
October 1991 (vol. 13 no. 10)
pp. 1050-1065

Results from an ongoing project concerned with recognizing objects in complex scene domains, especially in the domain that includes the natural outdoor world, are described. Traditional machine recognition paradigms assume either that all objects of interest are definable by a relatively small number of explicit shape models or that all objects of interest have characteristic, locally measurable features. The failure of both assumptions has a dramatic impact on the form of an acceptable architecture for an object recognition system. In this work, the use of the contextual information is a central issue, and a system is explicitly designed to identify and use context as an integral part of recognition that eliminates the traditional dependence on stored geometric models and universal image partitioning algorithms. This paradigm combines the results of many simple procedures that analyze monochrome, color, stereo, or 3D range images. Interpreting the results along with relevant contextual knowledge makes it possible to achieve a reliable recognition result, even when using imperfect visual procedures. Initial experimentation with the system on ground-level outdoor imagery has demonstrated competence beyond what is attainable with other vision systems.

[1] S. T. Barnard, "Stochastic stereo matching over scale,"Int. J. Comput. vision, vol. 3, no. 1, 1989.
[2] H. G. Barrow and J. M. Tenenbaum, "MSYS: A system for reasoning about scenes," Tech. Note 121, Artificial Intell. Cent., SRI Int., Apr. 1976.
[3] H. G. Barrow, T. D. Garvey, J. Kremers, J. M. Tenenbaum, and H. C. Wolf, "Interactive aids for cartography and interpretation," Tech. Note 137, Artificial Intell. Cent., SRI Int., Jan. 1977.
[4] R. C. Bolles, R. Horaud, and M. J. Hannah, "3DPO: A 3D part orientation system," inProc. 8th Int. Joint Conf. Artificial Intell.(Karlsruhe, West Germany), Aug. 1983, pp. 1116-1120.
[5] R. A. Brooks, "Model-based 3-D interpretations of 2-D images,"IEEE Trans. Patt. Anal. Machine Intell., vol. 5, no. 2, pp. 140-150, Mar. 1983.
[6] B. A. Draper, R. T. Collins, J. Brolio, A.R. Hanson, and E.M. Riseman, "The schema system,"Int. J. Comput. Vision, vol. 2, no. 3, pp. 209-250, Jan. 1989.
[7] M. A. Fischler and T. M. Strat, "Recognizing trees, bushes, rocks, and rivers," inProc. AAAI Spring Symp. Series: Phys. Biologic. Approaches Computation. Vision(Stanford Univ., Stanford, CA), Mar. 1988, pp. 62-64.
[8] M. A. Fischler and T. M. Strat, Recognizing objects in a natural environment: A contextual vision system," inProc. DARPA Image Understanding Workshop(Palo Alto, CA), May 1989, pp. 774-796.
[9] P. Fua and A. J. Hanson, "Using generic geometric models for intelligent shape extraction," inProc. DARPA Image Understanding Workshop(Los Angeles, CA), Feb. 1987, pp. 227-233.
[10] M. J. Hannah, "SRI's baseline stereo system," inProc. DARPA Image Understanding Workshop(Miami Beach, FL), Dec. 1985, pp. 149-155.
[11] A. R. Hanson and E. M. Riseman, "VISIONS: A computer system for interpreting scenes," inComputer Vision Systems. New York: Academic, 1978, pp. 303-333.
[12] A. J. Hanson and L. Quam, "Overview of the SRI cartographic modeling environment," inProc. DARPA Image Understanding Workshop(Cambridge, MA), Apr. 1988, pp. 576-582.
[13] D. P. Huttenlocher and S. Ulman, "Recognizing solid objects by alignment," inProc. DARPA Image Understanding Workshop(Cambridge, MA), Apr. 1988, pp. 1114-1122.
[14] K. I. Laws, "Integrated split/merge image segmentation," in Tech. Note 441, Artificial Intell. Cent., SRI Int., July 1988.
[15] D. M. McKeown, Jr., W. A. Harvey, Jr., and J. McDermott, "Rule-based interpretation of aerial imagery,"IEEE Trans. Patt. Anal. Machine Intell., vol. PAMI-7, no. 5, pp. 570-585, Sept. 1985.
[16] Y. Ohta, "A region-oriented image-analysis system by computer," Doctoral dissertation, Inform. Sci. Dept., Kyoto Univ., Kyoto, Japan, 1980.
[17] A. Rosenfeld, R. A. Hummel, and S. W. Zucker, "Scene labeling by relaxation operations,"IEEE Trans. Syst. Man Cybernetics, vol. SMC-6, no. 6, pp. 420-433, June 1976.
[18] K. R. Sloan, Jr., "World model driven recognition of natural scenes," Ph.D. Thesis, Moore School Elec. Eng., Univ. Pennsylvania, Philadelphia, June 1977.
[19] G. B. Smith and T. M. Strat, "Information management in a sensor-based autonomous system,"Proc. DARPA Image Understanding Workshop(Los Angeles, CA), Feb. 1987, pp. 170-177.
[20] T. M. Strat and G. B. Smith, "A knowledge-based information manager for autonomous vehicles," inImage Understanding in Unstructured Environments(S. Chen, Ed.). Cleveland, OH: World, 1988, pp. 1-39., ch. 1.
[21] T. M. Strat and M. A. Fischler, "Context-based vision: Recognition of natural scenes," inProc. 23rd Asilomar Conf. Signals Syst. Man, Oct. 1989, pp. 532-536.
[22] T. M. Strat, "Natural object recognition," Ph.D. Dissertation, Dept. Comput. Sci., Stanford Univ., Stanford, CA, Dec. 1990.
[23] J. M. Tenenbaum and S. Weyl, "A region analysis subsystem for interactive scene analysis," inProc. Fourth Int. Joint Conf. Artificial Intell.(Tbilisi, USSR), Sept. 1975, pp. 682-687.
[24] J. K. Tsotsos, "A complexity level analysis of immediate vision,"Int. J. Comput. Vision, vol. 1, no. 4, pp. 303-320, 1988.
[25] Y. Yakimovksy and J. A. Feldman, "A semantics-based decision theory region analyzer," inProc. Third Joint Conf. Artificial Intell.(Stanford, CA), Aug. 1973, pp. 580-588.

Index Terms:
context-based vision; object recognition; 2D imagery; complex scene domains; natural outdoor world; pattern recognition; picture processing
Citation:
T.M. Strat, M.A. Fischler, "Context-Based Vision: Recognizing Objects Using Information from Both 2D and 3D Imagery," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 10, pp. 1050-1065, Oct. 1991, doi:10.1109/34.99238
Usage of this product signifies your acceptance of the Terms of Use.