Data- and Model-Driven Multiresolution Processing

We introduce a technique formultiresolution processingwhich elegantly fits in our framework for visual recognition, described in earlier papers. The input is processedsimultaneouslyat a coarse resolution throughout the image and at finer resolution within a small window (fovea). We introduce an approach for controlling the movement of the high-resolution window which allows for both data- and model-driven selection of fixation points. Three fixation modes have been implemented, one based on large unexplained areas in the data, one on conflicts in the object-model database, and one on a 2D “space filling” algorithm. We argue that this kind of multiresolution processing is not only useful in limiting the computational time, as has been widely recognized, but also can be a deciding factor in making the entire vision problem a tractable and stable one. To demonstrate the approach, we introduce a class of3D surface texturesas a feature for recognition in our system. Surface texture recognition typically requires higher-resolution processing than that required for the extraction of the underlying surface. As examples, surface texture is used to discriminate between a ping-pong ball and a golf ball, and “curve texture” is used to recognize different types of gears. Other experimental results also are included to show the advantages and the implications of our approach.

[1]  Ruud M. Bolle,et al.  A Complete and Extendable Approach to Visual Recognition , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Andrea Califano,et al.  Data and model driven foveation , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[3]  Peter J. Burt,et al.  Smart sensing within a pyramid vision machine , 1988, Proc. IEEE.

[4]  Jerome A. Feldman,et al.  Connectionist Models and Their Properties , 1982, Cogn. Sci..

[5]  R. Bajcsy.  Active perception , 1988 .

[6]  Edith Schonberg,et al.  Two-Dimensional, Model-Based, Boundary Matching Using Footprints , 1986 .

[7]  Dana H. Ballard,et al.  Eye Fixation And Early Vision: Kinetic Depth , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[8]  Christopher M. Brown,et al.  Task-specific utility in a general Bayes net vision system , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[9]  Ruud M. Bolle,et al.  The Multiple Window Parameter Transform , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  R.M. Haralick,et al.  Statistical and structural approaches to texture , 1979, Proceedings of the IEEE.

[11]  H. Schuster.  Deterministic chaos: An introduction , 1984 .

[12]  Dana H. Ballard,et al.  Reference Frames for Animate Vision , 1989, IJCAI.

[13]  Thomas O. Binford,et al.  Survey of Model-Based Image Analysis Systems , 1982 .

[14]  Ramesh C. Jain,et al.  Three-dimensional object recognition , 1985, CSUR.

[15]  P. J. Burt,et al.  Fast Filter Transforms for Image Processing , 1981 .

[16]  Rakesh Mohan,et al.  Active 3D object models , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[17]  D. Noton,et al.  Eye movements and visual perception. , 1971, Scientific American.

[18]  A. L. I︠A︡rbus.  Eye Movements and Vision , 1967 .

[19]  Theodosios Pavlidis,et al.  A hierarchical data structure for picture processing , 1975 .

[20]  Luc Van Gool,et al.  Texture analysis Anno 1983 , 1985, Comput. Vis. Graph. Image Process..

[21]  Manfredo P. do Carmo,et al.  Differential geometry of curves and surfaces , 1976 .

[22]  C. Lin,et al.  Adaptive Moving Object Tracking Integrating Neural Networks And Intelligent Processing , 1989, Other Conferences.

[23]  Dana H. Ballard,et al.  Parameter Networks: Towards a Theory of Low-Level Vision , 1981, IJCAI.

[24]  Christos H. Papadimitriou,et al.  The complexity of recognizing polyhedral scenes , 1985, 26th Annual Symposium on Foundations of Computer Science (sfcs 1985).