Computer Vision through Learning

Abstract : The underlying motivation for this research is that vision systems need learning capabilities for handling problems for which algorithmic solutions are unknown or difficult to obtain. In this context, we have conducted research on a wide range of vision problems that can benefit from machine learning. We have developed a general methodology for this purpose, called MIST, that supports multilevel image sampling, transformation, learning and interpretation. MIST is based on the application of symbolic or multistrategy learning methods (the latter one integrates symbolic and neural net learning) for creating visual concept descriptions. MIST was applied and demonstrated to be useful for such problems as conceptual interpretation of natural scences, non-structural and structural texture description and identification, detection of blasting caps in X-ray images of airport luggage, and target detection in SAR images. We have also obtained important results in the areas of action recognition in video image sequences, visual memories, estimation of environment properties from sampling, and bisight head control. The obtained results have demonstrated a significant promise and usefulness of the efforts to apply modern machine learning methods to problems of computer vision.

[1]  Jerzy W. Bala,et al.  The Principal Axes Method for Constructive Induction , 1992, ML.

[2]  Thomas M. Strat,et al.  Natural Object Recognition: A Theoretical Framework and Its Implementation , 1991, IJCAI.

[3]  Diane J. Cook,et al.  Learning Membership Functions in a Function-Based Object Recognition System , 1995, J. Artif. Intell. Res..

[4]  B. A. Shepherd,et al.  An Appraisal of a Decision Tree Approach to Image Classification , 1983, IJCAI.

[5]  Kevin W. Bowyer,et al.  Indexing function-based categories for generic recognition , 1992, CVPR.

[6]  Michael R. Lowry,et al.  Learning Physical Descriptions From Functional Definitions, Examples, and Precedents , 1983, AAAI.

[7]  Ryszard S. Michalski,et al.  THE PRAX APPROACH TO LEARNING A LARGE NUMBER OF TEXTURE CONCEPTS , 1993 .

[8]  Casimir A. Kulikowski,et al.  Computer Systems That Learn: Classification and Prediction Methods from Statistics, Neural Nets, Machine Learning and Expert Systems , 1990 .

[9]  Ruzena Bajcsy,et al.  Shape and Function , 1987, Other Conferences.

[10]  Scott E. Fahlman,et al.  An empirical study of learning speed in back-propagation networks , 1988 .

[11]  L. Stark,et al.  Dissertation Abstract , 1994, Journal of Cognitive Education and Psychology.

[12]  Ryszard S. Michalski,et al.  Learning Descriptions of 2D Blob-Like Shapes for Object Recognition in X-Ray Images: An Initial Study , 1994 .

[13]  Azriel Rosenfeld,et al.  Machine Vision and Learning: Research Issues and Directions , 1994 .

[14]  Kevin W. Bowyer,et al.  Generic recognition of articulated objects by reasoning about functionality , 1994, ICPR.

[15]  Azriel Rosenfeld,et al.  Recognition by Functional Parts , 1995, Comput. Vis. Image Underst..

[16]  Tomaso A. Poggio,et al.  Example-Based Learning for View-Based Human Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Ruzena Bajcsy,et al.  Active Investigation of Functionality , 1994 .

[18]  Stanley M. Dunn,et al.  Learning Shape Classes , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[19]  Dean Pomerleau,et al.  Efficient Training of Artificial Neural Networks for Autonomous Navigation , 1991, Neural Computation.

[20]  Kenneth A. Kaufman,et al.  AQ17 A Multistrategy Learning System The Method and Users Guide , 1993 .

[21]  Ryszard S. Michalski,et al.  Pattern Recognition as Rule-Guided Inductive Inference , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[22]  Azriel Rosenfeld,et al.  Learning in Navigation Goal Finding in Graphs , 1996, Int. J. Pattern Recognit. Artif. Intell..

[23]  Ryszard S. Michalski,et al.  A Variable-Valued Logic System as Applied to Picture Description and Recognition , 1972 .

[24]  Jacek M. Zurada,et al.  Introduction to artificial neural systems , 1992 .

[25]  Ehud Rivlin,et al.  Function From Motion , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[26]  D R Dance.  Diagnostic Radiology with X-Rays , 1988 .

[27]  Dmitry B. Goldgof,et al.  Function-based recognition from incomplete knowledge of shape , 1993 .

[28]  Nada Lavrac,et al.  The Multi-Purpose Incremental Learning System AQ15 and Its Testing Application to Three Medical Domains , 1986, AAAI.

[29]  Avinash C. Kak,et al.  Automatic Generation of Object Class Descriptions Using Symbolic Learning Techniques , 1991, AAAI.

[30]  Allen Newell,et al.  A Model for Functional Reasoning in Design , 1971, IJCAI.

[31]  A. Rosenfeld,et al.  PROGRESS ON VISION THROUGH LEARNING † : A Collaborative Effort of George Mason University and University of Maryland , 1996 .

[32]  Peter W. Pachowicz,et al.  Progress on Vision Through Learning at George Mason University , 1994 .

[33]  Peter Bock,et al.  Using a hierarchical approach to avoid over-fitting in early vision , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[34]  Takeo Kanade,et al.  Neural Network-Based Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[35]  Michael Brady,et al.  Generating and Generalizing Models of Visual Objects , 1987, Artif. Intell..

[36]  Azriel Rosenfeld,et al.  Recognizing Blasting Caps in X-Ray Images , 1996 .

[37]  Kevin W. Bowyer,et al.  Generic recognition through qualitative reasoning about 3-D shape and object function , 1991, CVPR.

[38]  Ryszard S. Michalski,et al.  AQVAL/1--Computer Implementation of a Variable-Valued Logic System VL1 and Examples of its Application to Pattern Recognition , 1973, IJCAI 1973.

[39]  Kenneth A. Kaufman,et al.  Inductive Learning System AQ15c: The Method and User's Guide , 1995 .

[40]  I. Biederman.  Human image understanding: Recent research and a theory , 1985, Computer Vision Graphics and Image Processing.