A computational measure of saliency of the shape of 3D objects

A computational model of perceptual saliency, aimed to the selection of a reference object for locating objects within a virtual environment, is presented. The basic features of a 3D object (i.e. color, size and shape) are individually extracted through different proposed methods; this information is then normalized and the values are combined to get the individual saliency for each 3D object within the virtual environment. The main contribution of our computational model is to provide a measure of perceptual saliency of 3D objects, which considers the object?s features with the major stimulus for the human visual system