DEVELOPMENT OF LOCALIZED ORIENTED RECEPTIVE-FIELDS BY LEARNING A TRANSLATION-INVARIANT CODE FOR NATURAL IMAGES

Neurons in the mammalian primary visual cortex are known to possess sp atially localized, oriented receptive fields. It has previously been s uggested that these distinctive properties may reflect an efficient im age encoding strategy based on maximizing the sparseness of the distri bution of output neuronal activities or alternately, extracting the in dependent components of natural image ensembles. Here, we show that a strategy for transformation-invariant coding of images based on a firs t-order Taylor series expansion of an image also causes localized, ori ented receptive fields to be learned from natural image inputs. These receptive fields, which approximate localized first-order differential operators at various orientations, allow a pair of cooperating neural networks, one estimating object identity ('what') and the other estim ating object transformations ('where'), to simultaneously recognize an object and estimate its pose by jointly maximizing the a posteriori p robability of generating the observed visual data. We provide experime ntal results demonstrating the ability of such networks to factor reti nal stimuli into object-centred features and object-invariant transfor mation estimates.