We model the neural dynamics in the primate primary visual cortex in terms of a continuous director field that describes the average rate and the average orientational preference of active neurons at a particular point in the cortex. This representation has a correspondence to the Landau - de Gennes order parameter for nematic liquid crystal in two dimensions. Our linear-nonlinear dynamical model incorporates long range connectivity patterns that enforce context dependence present in the visual field. The model can distinguish large contiguous objects from the background clutter by suppressing the clutter and by filling-in occluded elements of object contours. This results in high-precision, high-recall detection of large objects in cluttered scenes.
CITATION STYLE
Singh, V., Tchernookov, M., Butterfield, R., & Nemenman, I. (2013). Continuum dynamics model of the primary visual cortex for contour detection, 7. Retrieved from http://arxiv.org/abs/1310.1341
Mendeley helps you to discover research relevant for your work.