Self-Organization of Steerable Topographic Mappings as Basis for Translation Invariance

One way to handle the perception of images that change in position (or size, orientation or deformation) is to invoke rapidly changing fiber projections to project images into a fixed format in a higher cortical area. We propose here a model for the ontogenesis of the necessary control structures. For simplicity we limit ourselves to fiber projections between two one-dimensional chains of units. Our system is a direct extension of a mathematical model [1] for the ontogenesis of retinotopy. Our computer experiments are guided by stability analysis and show the establishment of multiple topographic mappings implementing different translations, each projection associated with a single control unit. The model relies on neural signals with appropriate correlation structure, signals that can be generated by the network as spontaneous noise, so that the proposed mechanism could act prenatally.

[1]  Geoffrey E. Hinton A Parallel Computation that Assigns Canonical Object-Based Frames of Reference , 1981, IJCAI.

[2]  Christoph von der Malsburg,et al.  Self-Organization of Topographic Bilinear Networks for Invariant Recognition , 2011, Neural Computation.

[3]  Rajesh P. N. Rao,et al.  Bilinear Sparse Coding for Invariant Vision , 2005, Neural Computation.

[4]  Rajesh P. N. Rao,et al.  Learning Lie Groups for Invariant Visual Perception , 1998, NIPS.

[5]  O. G. Selfridge,et al.  Pandemonium: a paradigm for learning , 1988 .

[6]  G. Goodhill Contributions of Theoretical Modeling to the Understanding of Neural Map Development , 2007, Neuron.

[7]  Bruno A. Olshausen,et al.  Bilinear models of natural images , 2007, Electronic Imaging.

[8]  Christoph von der Malsburg,et al.  Maplets for correspondence-based object recognition , 2004, Neural Networks.

[9]  Johan Eriksson,et al.  Looking as if you know: Systematic object inspection precedes object recognition. , 2008, Journal of vision.

[10]  Christian Wolff,et al.  A recurrent dynamic model for correspondence-based face recognition. , 2008, Journal of vision.

[11]  Tomaso Poggio,et al.  Models of object recognition , 2000, Nature Neuroscience.

[12]  Christoph von der Malsburg Network Self-Organization in the Ontogenesis of the Mammalian Visual System , .

[13]  Jörg Lücke,et al.  Rapid Convergence to Feature Layer Correspondences , 2008, Neural Computation.

[14]  D J Willshaw,et al.  A marker induction mechanism for the establishment of ordered neural mappings: its application to the retinotectal problem. , 1979, Philosophical transactions of the Royal Society of London. Series B, Biological sciences.

[15]  Takayuki Ito,et al.  Neocognitron: A neural network model for a mechanism of visual pattern recognition , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[16]  Joshua B. Tenenbaum,et al.  Separating Style and Content with Bilinear Models , 2000, Neural Computation.

[17]  D. V. van Essen,et al.  A neurobiological model of visual attention and invariant pattern recognition based on dynamic routing of information , 1993, The Journal of neuroscience : the official journal of the Society for Neuroscience.

[18]  Joachim M. Buhmann,et al.  Distortion Invariant Object Recognition in the Dynamic Link Architecture , 1993, IEEE Trans. Computers.

[19]  Joel L. Davis,et al.  An Introduction to Neural and Electronic Networks , 1995 .

[20]  Geoffrey E. Hinton,et al.  Unsupervised Learning of Image Transformations , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  H. A.F.,et al.  DEVELOPMENT OF RETINOTOPIC PROJECTIONS: AN ANALYTIC TREATMENT , 1983 .