A Model of the Acquisition of Object Representations in Human 3D Visual Recognition

A common approach to the study of visual recognition postulates that there exist in the visual system representations of familiar objects and scenes. To recognize an object, the system compares it with each of the stored models. Such a comparison would appear possible only after the input image and the stored representations are brought to a common form. Consequently, the nature of representation must be reflected in the performance of the system [7]