Vision, analog networks, and the minimum norm constraint

Several analog networks have been proposed for solving the variational problems of early vision. The networks can be implemented in hardware using analog VLSI: thus, they can be used in real-time environments. The underlying mathematical formulations do not necessarily lead to a unique solution. The authors show how the nonuniqueness carries over to the analog networks. In particular, they study networks for contour-based optical flow, area-based optical flow, membrane surface reconstruction, and thin-plate surface reconstruction. An additional constraint, the minimum norm constraint, is proposed in the variational formulations of these problems. The minimum norm constraint ensures a unique solution. In analog networks, the constraint can be imposed simply by shunting each node to ground through an appropriate positive resistor. The minimum norm constraint also tends to force the solution to conform more with local measurements. This is in accord with a psychophysical study which has revealed the presence of such a tendency in the human visual system

[1]  Timothy A. Grogan,et al.  Depth From Stereo: Variational Theory And A Hybrid Analog-Digital Network , 1989, Photonics West - Lasers and Applications in Science and Engineering.

[2]  T. Poggio,et al.  III-Posed problems early vision: from computational theory to analogue networks , 1985, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[3]  Jin Luo,et al.  Computing motion using analog and binary resistive networks , 1988, Computer.

[4]  N M Grzywacx,et al.  Motion correspondence and analog networks , 1987 .

[5]  J. J. Hopfield,et al.  “Neural” computation of decisions in optimization problems , 1985, Biological Cybernetics.

[6]  David Lee,et al.  Computational Aspects Of Determining Optical Flow , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[7]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[8]  Carver Mead,et al.  Analog VLSI and neural systems , 1989 .

[9]  Christof Koch,et al.  Computing optical flow in resistive networks and in the primate visual system , 1989, [1989] Proceedings. Workshop on Visual Motion.

[10]  Ellen C. Hildreth,et al.  Measurement of Visual Motion , 1984 .

[11]  Christof Koch,et al.  Simple analog and hybrid networks for surface interpolation , 1987 .

[12]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[13]  Tomaso A. Poggio,et al.  Motion Field and Optical Flow: Qualitative Properties , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  J. Harris.  The Coupled Depth/Slope Approach to Surface Reconstruction , 1986 .

[15]  K. Nakayama,et al.  The aperture problem—II. Spatial integration of velocity information along contours , 1988, Vision Research.

[16]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.

[17]  John G. Harris,et al.  An Analog VLSI Chip for Thin-Plate Surface Interpolation , 1988, NIPS.

[18]  Timothy A. Grogan,et al.  Uniqueness, the minimum norm constraint, and analog networks for optical flow along contours , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[19]  Eric L. W. Grimson,et al.  From Images to Surfaces: A Computational Study of the Human Early Visual System , 1981 .

[20]  John L. Wyatt,et al.  Criteria for Robust Stability In A Class Of Lateral Inhibition Networks Coupled Through Resistive Grids , 1989, Neural Computation.

[21]  C Koch,et al.  Analog "neuronal" networks in early vision. , 1986, Proceedings of the National Academy of Sciences of the United States of America.