Professor Aurel A. Lazar has received a $750,000 award from AFOSR for three years to investigate and develop representations of visual information using architectures inspired by the human visual system. The title of the grant is "The Geometry of Switching Architectures for Time Encoding".
Representations of the visual world in the brain are invariant with respect to a number of fluctuations occurring on the retina including object position, scale, pose and illumination, and the presence of clutter. Architectures for the representation and processing of information in the spike domain will fundamentally enable novel real-time object identification and recognition algorithms in the visual space.