Despite advances in artificial neural networks, extracting salient information from
images and reasoning about it like people do remains a challenge. This talk will describe
a new computational model of human primary visual cortex (visual area V1) that represents
the basic elements of shape in the same way human visual architecture does.