This Article 
 Bibliographic References 
 Add to: 
On the Use of SDF-Type Filters for Distortion Parameter Estimation
November 2002 (vol. 24 no. 11)
pp. 1521-1528

Abstract—Synthetic discriminant functions have been used to locate objects irrespective of distortions and to estimate the extent of the distortion. It was recognized from the beginning that accurate estimates are only possible provided the training set is constructed carefully. In this paper, we obtain conditions that will ensure the accuracy of the estimates. The conditions also suggest efficient ways of constructing the training sets and the results are extended to a wide class SDF-type filters. The theoretical results are illustrated with (idealized) examples and are also applied to the more realistic problem of accurate facial location.

[1] E. Barnard and D. Casasent, “Invariance and Neural Nets,” IEEE Trans. Neural Networks, vol. 2, no. 5, pp. 498-508, 1991.
[2] M.K. Bennett and P.K. Rajan, “Comparison of Five Types of Synthetic Estimation Filters for Pose Estimation,” Proc. SPIE, vol. 1702, pp. 121-131, 1992.
[3] C. de Boor, “Good Approximation by Splines with Variable Knots II,” Proc. Conf. Numerical Solution of Differential Equations, 1973.
[4] R.S. Embar and P.K. Rajan, “Design of Synthetic Estimation Filters Using Correlation Energy Minimization,” Proc. SPIE, vol. 2237, pp. 124-135, 1994.
[5] C.F. Hester and D. Casasent, “Multivariant Techniques for Multiclass Pattern Recognition,” Applied Optics, vol. 19, pp. 1758-1761, 1980.
[6] K. Hotta, T. Kurita, and T. Mishima, “Scale Invariant Face Detection Method Using Higher-Order Local Autocorrelation Features of Log-Polar Image,” Proc. Third IEEE Int'l Conf. Automatic Face and Gesture Recognition, pp. 70-75, 1998.
[7] K. Jonsson, J. Matas, and J. Kittler, “Learning Salient Features for Real-Time Face Verification,” Audio- and Video-Based Biometric Person Authentication, S. Akunuri and C. Kullman, eds., pp. 60-65, 1999.
[8] R.D. Juday and S.E. Monroe, “Synthetic Estimation Filters for a Control System,” Proc. SPIE, vol. 754, pp. 40-47, 1987.
[9] A. Kostrzewski, D.H. Kim, G. Savant, J. Kim, A. Vasiliev, and T. Jannson, “Highly Robust Pose Estimator Based on Synthetic Estimation Filters,” Proc. SPIE, vol. 3159, pp. 60-66, 1997.
[10] B.V.K.V. Kumar, J.D. Brasher, and C.F. Hester, “Role of Constraint Values in Synthetic Discriminant Filter Design,” Proc. SPIE, vol. 1959, pp. 23-31, 1993.
[11] B.V.K.V. Kumar and A. Mahalanobis, “Recent Advances in Distortion-Invariant Correlation Filter Design,” Proc. SPIE, vol. 2490, pp. 2-13, 1995.
[12] B.V.K.V. Kumar, A. Mahalanobis, S. Song, S.R.F. Sims, and J.F. Epperson, “Minimum Squared Error Synthetic Discriminant Functions,” Optical Eng., vol. 31, no. 5, pp. 915-922, May 1992.
[13] B.V.K.V. Kumar, A. Mahalanobis, and A. Takessian, “Optimal Tradeoff Circular Harmonic Function Correlation Methods Providing Controlled In-Plane Rotation Response,” IEEE Trans. Image Processing, vol. 9, pp. 1025-1034, 2000.
[14] A. Mahanobis, B.V.K.V. Kumar, S. Song, S.R.F. Sims, and J.F. Epperson, “Unconstrained Correlation Filters,” Applied Optics, vol. 33, pp. 3751-3759, 1994.
[15] B. Moghaddam and A. Pentland, "Probabilistic Visual Learning for Object Detection," Int'l Conf. Computer Vision, 1995, pp. 786-793.
[16] S.E. Monroe and R.D. Juday, “Extension of Synthetic Estimation Filters for Relative Position Measurements,” Proc. SPIE, vol. 938, pp. 253-260, 1988.
[17] S.E. Monroe and R.D. Juday, “Multidimensional Synthetic Estimation Filter,” Proc. SPIE, vol. 1347, pp. 179-185, 1990.
[18] N. Muller, “Facial Recognition, Eigenfaces and Synthetic Discriminant Functions,” PhD thesis, Univ. of Stellenbosch, 2000.
[19] N. Muller and B. Herbst, “On the Use of SDF-Type Filters for Distortion Invariant Image Location” Proc. 15th Int'l Conf. Pattern Recognition, vol. 3, pp. 530-533, 2000.
[20] L. Neiburg, “Feature Space Trajectory Pattern Classifier,” PhD thesis, Carnegie Mellon Univ., 1996.
[21] V. Pereya and E.G. Sewell, “Mesh Selection for Discrete Solution of Boundary Problems in Ordinary Differential Equations,” Numerical Math., vol. 23, pp. 261-268, 1975.
[22] G. Ravichandran and D. Casasent, “Minimum Noise and Correlation Energy Optical Correlation Filter,” Applied Optics, vol. 31, no. 11, Apr. 1992.
[23] P. Réfrégier and V. Laude, “Critical Analysis of Filtering Techniques for Optical Pattern Recognition: Are the Solutions of this Inverse Problem Stable,” Proc. SPIE, vol. 12, pp. 58-84, 1994.
[24] H. Rowley, S. Baluja, and T. Kanade, "Neural Network-Based Face Detection," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 20, no. 1, Jan. 1998, pp. 23-38.
[25] P. Simard, Y. Le Cun, J. Denker, and B. Victorri, “Transformation Invariance in Pattern Recognition-Tangent Distance and Tangent Propagation,” Lecture Notes in Computer Science, vol. 1524, pp. 239-274, 1998.
[26] P. Simard, B. Victorri, Y. Le Cun, and J. Denker, “Tangent Prop—A Formalism for Specifying Selected Invariances in an Adaptive Network,” Advances in Neural Information Processings Systems, J.E. Moody, S.J. Hanson, and R.P. Lippmann, eds., pp. 651-655, 1992.
[27] M.A. Sipe, ”Feature Space Trajectory Methods for Active Object Recognition,” PhD thesis, Carnegie Mellon Univ., 1999.
[28] G. Strang, Introduction to Linear Algebra, second ed. Wellesley-Cambridge Press, 1998.

Index Terms:
Synthetic discriminant functions, synthetic estimation filters, facial location.
Neil Muller, B.M. Herbst, "On the Use of SDF-Type Filters for Distortion Parameter Estimation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 11, pp. 1521-1528, Nov. 2002, doi:10.1109/TPAMI.2002.1046173
Usage of this product signifies your acceptance of the Terms of Use.