Representing 3D objects by sets of activities of receptive fields

Idealized models of receptive elds (RFs) can be used as building blocks for the creation of powerful distributed computation systems. The present report concentrates on investigating the utility of collections of RFs in representing 3D objects under changing viewing conditions. The main requirement in this task is that the pattern of activity of RFs vary as little as possible when the object and the camera move relative to each other. I propose a method for representing objects by RF activities, based on the observation that, in the case of rotation around a xed axis, dierences of activities of RFs that are properly situated with respect to that axis remain invariant. Results of computational experiments suggest that a representation scheme based on this algorithm for the choice of stable pairs of RFs would perform consistently better than a scheme involving random sets of RFs. The proposed scheme may be useful under object or camera rotation, both for ideal Lambertian objects, and for real-world objects such as human faces. 1 Introduction Many of the lower-level areas in the primate visual system are organized retinotopically, that is, as maps which preserve to a certain degree the topography of the retina. A unit that is a part of such a retinotopic map normally responds selectively to stimulation in a well-localized part of the visual eld, referred to as its receptive eld (RF). Dierent regions within the RF may contribute dierently to the activity of the unit, according to the prole or the weighting function of the RF. The activity of the unit is frequently modeled by a (possibly nonlinear) function of the convolution of the activity distribution over the input area with the RF prole.

[1]  D. Hubel,et al.  Receptive fields of single neurones in the cat's striate cortex , 1959, The Journal of physiology.

[2]  John Y. Aloimonos,et al.  Unification and integration of visual modules: an extension of the Marr Paradigm , 1989 .

[3]  T Poggio,et al.  Regularization Algorithms for Learning That Are Equivalent to Multilayer Networks , 1990, Science.

[4]  Nathan Intrator,et al.  Objective function formulation of the BCM theory of visual cortical plasticity: Statistical connections, stability conditions , 1992, Neural Networks.

[5]  S Amari,et al.  A theory on the determination of 3D motion and 3D structure from features. , 1987, Spatial vision.

[6]  C S Harris,et al.  Is Orentation-Specific Color Adaptation in Human Vision Due to Edge Detectors, Afterimages, or "Dipoles"? , 1968, Science.

[7]  Yiannis Aloimonos,et al.  Active vision , 2004, International Journal of Computer Vision.

[8]  C. McCollough.  Color Adaptation of Edge-Detectors in the Human Visual System , 1965, Science.

[9]  T. Poggio.  A theory of how the brain might work. , 1990, Cold Spring Harbor symposia on quantitative biology.

[10]  Andrew Zisserman,et al.  Geometric invariance in computer vision , 1992 .

[11]  James F. Blinn,et al.  Models of light reflection for computer synthesized pictures , 1977, SIGGRAPH.

[12]  Shimon Edelman,et al.  Learning to Recognize Faces from Examples , 1992, ECCV.

[13]  S. Ullman,et al.  The interpretation of visual motion , 1977 .

[14]  H. Spitzer,et al.  Complex-cell receptive field models , 1988, Progress in Neurobiology.

[15]  T Poggio,et al.  Fast perceptual learning in visual hyperacuity. , 1991, Science.

[16]  S. W. Kuffler.  Discharge patterns and functional organization of mammalian retina. , 1953, Journal of neurophysiology.

[17]  P. O. Bishop,et al.  Receptive fields of simple cells in the cat striate cortex , 1973, The Journal of physiology.

[18]  K. Kanatani.  Group-Theoretical Methods in Image Understanding , 1990 .

[19]  K. Torrance,et al.  Polarization, Directional Distribution, and Off-Specular Peak Phenomena in Light Reflected from Roughened Surfaces , 1966 .

[20]  Heinrich H. Bülthoff,et al.  Psychophysical support for a 2D view interpolation theory of object recognition , 1991 .

[21]  Hanspeter A. Mallot,et al.  Neural mapping and space-variant image processing , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[22]  Tomaso Poggio,et al.  Optical flow: computational properties and networks, biological and analog , 1989 .

[23]  L C Katz,et al.  Development of local circuits in mammalian visual cortex. , 1992, Annual review of neuroscience.

[24]  Nathan Intrator.  Feature Extraction using an Unsupervised Neural Network , 1991 .