Autonomous Learning of Object Representations Utilizing Self { Controlled Movements 1

We introduce an object recognition system based on bent and stretched Gabor wavelets, called banana wavelets. Banana wavelets can be metrically organized. Utilizing this metric, representations of objects are learned autonomously on training data generated by a robot arm combined with a camera. We discuss possible analogies to learning of object representations by infants.

[1]  T G Bower,et al.  The object in the world of the infant. , 1971, Scientific American.

[2]  Robert M. Gray,et al.  An Algorithm for Vector Quantizer Design , 1980, IEEE Trans. Commun..

[3]  Elie Bienenstock,et al.  Neural Networks and the Bias/Variance Dilemma , 1992, Neural Computation.

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

[5]  David I. Perrett,et al.  Modeling visual recognition from neurobiological constraints , 1994, Neural Networks.

[6]  David J. Field,et al.  What Is the Goal of Sensory Coding? , 1994, Neural Computation.