A self-organizing multiple-view representation of 3D objects

We explore representation of 3D objects in which several distinct 2D views are stored for each object. We demonstrate the ability of a two-layer network of thresholded summation units to support such representations. Using unsupervised Hebbian relaxation, the network learned to recognize ten objects from different viewpoints. The training process led to the emergence of compact representations of the specific input views. When tested on novel views of the same objects, the network exhibited a substantial generalization capability. In simulated psychophysical experiments, the network's behavior was qualitatively similar to that of human subjects.

[1]  Christof Koch,et al.  Selecting One Among the Many: A Simple Network Implementing Shifts in Selective Visual Attention , 1984 .

[2]  Shimon Edelman,et al.  Viewpoint-specific Representations in Three-dimensional Object Recognition , 1990 .

[3]  Antonio R. Damasio,et al.  The Brain Binds Entities and Events by Multiregional Activation from Convergence Zones , 1989, Neural Computation.

[4]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[5]  S. Edelman,et al.  Stimulus Familiarity Determines Recognition Strategy for Novel 3D Objects , 1989 .

[6]  A. J. Mistlin,et al.  Visual neurones responsive to faces , 1987, Trends in Neurosciences.

[7]  John J. L. Morton,et al.  Interaction of information in word recognition. , 1969 .

[8]  M. Tarr,et al.  Mental rotation and orientation-dependence in shape recognition , 1989, Cognitive Psychology.

[9]  Takeo Kanade,et al.  Applying Sensor Models To Automatic Generation Of Object Recognition Programs , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

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

[11]  I. Rock,et al.  A case of viewer-centered object perception , 1987, Cognitive Psychology.

[12]  H. C. Longuet-Higgins,et al.  A computer algorithm for reconstructing a scene from two projections , 1981, Nature.

[13]  W. J. Nowack.  Neurobiology of Neocortex , 1989, Neurology.

[14]  Jerome A. Feldman,et al.  Connectionist Models and Their Properties , 1982, Cogn. Sci..

[15]  S. Ullman.  Aligning pictorial descriptions: An approach to object recognition , 1989, Cognition.

[16]  D. W. Thompson,et al.  Three-dimensional model matching from an unconstrained viewpoint , 1987, Proceedings. 1987 IEEE International Conference on Robotics and Automation.

[17]  David G. Lowe,et al.  Perceptual Organization and Visual Recognition , 2012 .

[18]  T. Poggio,et al.  A network that learns to recognize three-dimensional objects , 1990, Nature.

[19]  B. McNaughton,et al.  Hippocampal synaptic enhancement and information storage within a distributed memory system , 1987, Trends in Neurosciences.

[20]  S Ullman,et al.  Shifts in selective visual attention: towards the underlying neural circuitry. , 1985, Human neurobiology.

[21]  Alan L. Yuille,et al.  A Winner-Take-All Mechanism Based on Presynaptic Inhibition Feedback , 1989, Neural Computation.

[22]  R. Shepard,et al.  Mental Rotation of Three-Dimensional Objects , 1971, Science.

[23]  Massimo Piattelli-Palmarini,et al.  Evolution, selection and cognition: From “learning” to parameter setting in biology and in the study of language , 1989, Cognition.

[24]  Ronen Basri,et al.  Recognition by Linear Combinations of Models , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[25]  M. Alexander,et al.  Principles of Neural Science , 1981 .

[26]  John H. Holland,et al.  Escaping brittleness: the possibilities of general-purpose learning algorithms applied to parallel rule-based systems , 1995 .

[27]  Shimon Ullman,et al.  An Approach to Object Recognition: Aligning Pictorial Descriptions , 1986 .

[28]  Thomas S. Huang,et al.  Uniqueness and Estimation of Three-Dimensional Motion Parameters of Rigid Objects with Curved Surfaces , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[30]  A Koriat,et al.  Mental rotation and visual familiarity , 1985, Perception & psychophysics.

[31]  D. O. Hebb,et al.  The organization of behavior , 1988 .

[32]  Daphna Weinshall,et al.  Exploring Representation of 3D Objects for Visual Recognition , 1989 .

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

[34]  David H. Foster.  A hypothesis connecting visual pattern recognition and apparent motion , 2004, Kybernetik.

[35]  A. Larsen,et al.  Pattern matching: Effects of size ratio, angular difference in orientation, and familiarity , 1985, Perception & psychophysics.

[36]  Daphna Weinshall,et al.  A Self-Organizing Multiple-View Representation of Three- Dimensional Objects , 1989 .

[37]  R. Shepard,et al.  Mental Images and Their Transformations , 1982 .

[38]  Tomaso A. Poggio,et al.  Extensions of a Theory of Networks for Approximation and Learning , 1990, NIPS.

[39]  W. McCulloch.  Brain and Behavior. , 1951 .

[40]  I. Rock,et al.  Can we imagine how objects look from other viewpoints? , 1989, Cognitive Psychology.

[41]  Ralph Linsker,et al.  Self-organization in a perceptual network , 1988, Computer.

[42]  P. Jolicoeur.  The time to name disoriented natural objects , 1985, Memory & cognition.

[43]  Kunihiko Fukushima,et al.  Neocognitron: A hierarchical neural network capable of visual pattern recognition , 1988, Neural Networks.

[44]  H. Barlow.  Cerebral Cortex as Model Builder , 1987 .

[45]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[46]  James J. Little,et al.  Neural Architecture for Optical Flow Computation , 1987 .