An analog network for continuous-time segmentation

A common goal in computer vision is to segment scenes into different objects sharing common properties such as depth, motion, or image intensity. A segmentation algorithm has been developed utilizing an absolute-value smoothness penalty instead of the more common quadratic regularizer. This functional imposes a piece-wide constant constraint on the segmented data. Since the minimized energy is guaranteed to be convex, there are no problems with local minima, and no complex continuation methods are necessary to find the unique global minimum. This is in sharp contrast to previous software and hardware solutions to this problem. The energy minimized can be interpreted as the generalized power (or co-content) of a nonlinear resistive network. The network is called thetiny-tanh network since the I-V characteristic of the nonlinear resistor must be an extremely narrow-width hyperbolic-tangent function. This network has been demonstrated for 1-D step-edges with analog CMOS hardware and for a 2-D stereo algorithm in simulations.

[1]  James J. Little,et al.  Parallel Optical Flow Using Local Voting , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[2]  P. Green.  Bayesian reconstructions from emission tomography data using a modified EM algorithm. , 1990, IEEE transactions on medical imaging.

[3]  Christof Koch,et al.  Seeing Chips: Analog VLSI Circuits for Computer Vision , 1989, Neural Computation.

[4]  F. Girosi.  Models of Noise and Robust Estimates , 1991 .

[5]  Philip E. Gill,et al.  Practical optimization , 1981 .

[6]  Julian Besag,et al.  Digital Image Processing: Towards Bayesian image analysis , 1989 .

[7]  Massimo A. Sivilotti,et al.  Real-time visual computations using analog CMOS processing arrays , 1987 .

[8]  M. Bertero,et al.  Ill-posed problems in early vision , 1988, Proc. IEEE.

[9]  T. Poggio,et al.  Ill-posed problems in early vision: from computational theory to analogue networks , 1985, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[10]  Julian Besag,et al.  Towards Bayesian image analysis , 1993 .

[11]  Andrew Blake,et al.  Visual Reconstruction , 1987, Deep Learning for EEG-Based Brain–Computer Interfaces.

[12]  D. Shulman,et al.  Regularization of discontinuous flow fields , 1989, [1989] Proceedings. Workshop on Visual Motion.

[13]  J. G. Harris,et al.  Discarding outliers using a nonlinear resistive network , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[14]  John C. Platt,et al.  Constraint methods for neural networks and computer graphics , 1990 .

[15]  J. Harris.  Analog models for early vision , 1991 .

[16]  Christof Koch,et al.  Optical Flow and Surface Interpolation in Resistive Networks: Algorithms and Analog VLSI Chips , 1989 .

[17]  Brian G. Schunck,et al.  Robust computational vision , 1993, Other Conferences.

[18]  Richard F. Lyon,et al.  An analog electronic cochlea , 1988, IEEE Trans. Acoust. Speech Signal Process..

[19]  Leon O. Chua,et al.  Linear and nonlinear circuits , 1987 .

[20]  Donald Geman,et al.  Constrained Restoration and the Recovery of Discontinuities , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[21]  John Law,et al.  Robust Statistics—The Approach Based on Influence Functions , 1986 .

[22]  Brian G. Schunck.  Robust computational vision , 1995 .

[23]  C Koch,et al.  A two-dimensional analog VLSI circuit for detecting discontinuities in early vision. , 1990, Science.

[24]  John Lazzaro,et al.  A Silicon Model Of Auditory Localization , 1989, Neural Computation.

[25]  Misha Mahowald,et al.  A silicon model of early visual processing , 1993, Neural Networks.

[26]  Leon Levine,et al.  Methods for Solving Engineering Problems, Using Analog Computers , 1964 .

[27]  F. Girosi.  Models of Noise and Robust Estimation , 1991 .

[28]  Berthold K. P. Horn.  Parallel networks for machine vision , 1991 .

[29]  Tomaso Poggio,et al.  Probabilistic Solution of Ill-Posed Problems in Computational Vision , 1987 .

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

[31]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[32]  Jin Luo,et al.  Resistive Fuses: Analog Hardware for Detecting Discontinuities in Early Vision , 1989, Analog VLSI Implementation of Neural Systems.

[33]  Tomaso Poggio,et al.  An Analog Model of Computation for the Ill-Posed Problems of Early Vision, , 1984 .

[34]  Carver A. Mead.  Analog VLSI and neural systems (invited presentation) , 1989 .

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

[36]  B. Ripley,et al.  Robust Statistics , 2018, Wiley Series in Probability and Statistics.

[37]  A. Lumsdaine,et al.  Nonlinear analog networks for image smoothing and segmentation , 1990, IEEE International Symposium on Circuits and Systems.

[38]  Berthold K. P. Horn,et al.  Determining lightness from an image , 1974, Comput. Graph. Image Process..

[39]  W. Millar.  CXVI. Some general theorems for non-linear systems possessing resistance , 1951 .

[40]  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.