This Article 
 Bibliographic References 
 Add to: 
Robot Vision Using a Feature Search Strategy Generated from a 3D Oobject Model
October 1991 (vol. 13 no. 10)
pp. 1085-1097

A robot vision system that automatically generates an object recognition strategy from a 3D model and recognizes the object using this strategy is presented. The appearance of an object from various viewpoints is described in terms of visible 2D features such as parallel lines and ellipses. Features are then ranked according to the number of viewpoints from which they are visible. The rank and feature extraction cost of each feature are used to generate a treelike strategy graph. This graph gives an efficient feature search order when the viewpoint is unknown, starting with commonly occurring features and ending with features specific to a certain viewpoint. The system searches for features in the order indicated by the graph. After detection, the system compares a lines representation generated from the 3D model with the image features to localize the object. Perspective projection is used in the localization process to obtain the precise position and attitude of the object, whereas orthographic projection is used in the strategy generation process to allow symbolic manipulation. Experimental results are given.

[1] R. A. Brooks, "Symbolic reasoning among 3-D models and 2-D images,"Artificial Intell., vol. 17, pp. 285-348, Aug. 1981.
[2] C. Goad, "Special purpose automatic programming for 3D model-based vision," inProc. Image Understanding Workshop(Arlington, VA), June 1983, pp. 94-104.
[3] C.K. Cowan, D.M. Chelberg, and H.S. Lim, "ACRONYM model-based vision in the intelligent task automation project," inProc. AI Applications(Denver, CO), Dec. 1984, pp. 176-183.
[4] K. Ikeuchi and T. Kanade, "Automatic generation of object recognition programs,"Proc. IEEE, vol. 76, no. 8, pp. 1016-1035, Aug. 1988.
[5] K.S. Hong and K. Ikeuchi, "Generating a strategy for configuration determination: A module of vision algorithm compiler for object localization programs," inPreprints 5th Int. Symp. Robotics Res.(Tokyo), Aug. 1989, pp. 164-171.
[6] Y. Lamdan and H. J. Wolfson, "Geometric hashing: A general and efficient model-based recognition scheme," inProc. 2nd Int. Conf. Computer vision, 1988.
[7] C. Hansen and T. Henderson, "Towards the automatic generation of recognition strategies," inProc. 2nd ICCV(Tampa, FL), Dec. 1988, pp. 275-279.
[8] R. C. Bolles and R. A. Cain, "Recognizing and locating partially visible objects: The local-feature-focus method,"Int. J. Robotics Res., vol. 1, no. 3, pp. 57-82, 1982.
[9] R. C. Bolles and P. Horaud, "3DPO: A three dimensional part orientation svstem,"Int. J. Robotics Res., vol. 5, no. 3, Fall 1986, pp. 3-26.
[10] W. E. L. Grimson and T. Lozano-Pérez, "Model-based recognition and localization from sparse range or tactile data,"Int. J. Robotics Res., vol. 3, no. 3, pp. 3-35, 1984.
[11] D. Shoham and U. Ullman, "Aligning a model to an image using minimal information," inProc. 2nd ICCV Conf., 1988, pp. 259-263.
[12] D. G. Lowe, "Three-dimensional object recognition from single two-dimensional images,"Artificial Intell., vol. 31, 1987.
[13] D.G. Lowe, "The viewpoint consistency constraint,"Int. J. Comput. Vision, vol. 1, no. 1, pp. 57-72, 1987.
[14] D. Marr and H.K. Nishihara, "Visual information processing: Artificial intelligence and the sensorium of sight,"Technol. Rev., vol. 81, pp. 2-23, Oct. 1978.
[15] Y. Okamoto, Y. Kuno, K. Onoguchi, M. Watanabe, and H. Asada, "Object recognition using a tree-like procedure generated from 3-D model," inProc. IAPR Workshop Comput. Vision(Tokyo), Oct. 1988, pp. 441-446.
[16] R. Horaud, "New methods for matching 3-D objects with single perspective views,"IEEE Trans. Patt. Anal. Machine Intell., vol. PAMI-9, pp. 401-412, May 1987.
[17] M. Dhome, M. Richetin, J.T. Lapresté, and G. Rives, "Determination of the attitude of 3-D objects from a single perspective view,"IEEE Trans. Patt. Anal. Machine Intell., vol. 11, pp. 1265-1278, Dec. 1989.
[18] J. Ponce and D. J. Kriegman, "On recognizing and positioning curved 3D objects from image contours," inProc. IEEE Workshop on Interpretation of 3D Scenes, Austin, TX, Nov. 1989, pp. 61-67.
[19] Y. Okamoto, Y. Kuno, and S. Okada, "Robot vision using object models for recognition strategy generation and precise localization," inProc. IECON '90 16th Ann. Conf. IEEE Ind. Electron. Soc.(Pacific Grove, CA), Nov. 1990, pp. 558-563.
[20] M. Ishikawa and H. Hoshino, "High performance image processor for robot vision," inSpring Nat. Conv. Rec., Inst. Electron, Inform. Commun. Eng. Japan(Tokyo), Mar. 1988, p. 2-187 (in Japanese).
[21] R. Bajcsy, "Active perception,"Proc. IEEE, vol. 76, pp. 996-1005, Aug. 1988.

Index Terms:
object localization; perspective projection; position measurement; attitude determination; feature search strategy; 3D object model; robot vision; parallel lines; ellipses; feature extraction; treelike strategy graph; orthographic projection; strategy generation; symbolic manipulation; computer vision; computerised pattern recognition; computerised picture processing; robots
Y. Kuno, Y. Okamoto, S. Okada, "Robot Vision Using a Feature Search Strategy Generated from a 3D Oobject Model," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 10, pp. 1085-1097, Oct. 1991, doi:10.1109/34.99241
Usage of this product signifies your acceptance of the Terms of Use.