Recognition of Occluded Shapes Using Size Functions

The robustness against occlusions and the ability to performnot only global matching, but also partial matching are investigated incomputer vision in order to evaluate the performance of shape descriptors.In this paper we consider the size function shape descriptor, andwe illustrate some results about size functions of occluded shapes. Theoreticalresults indicate that size functions are able to detect a partialmatching between shapes by showing a common subset of cornerpoints.Experiments are presented which outline the potential of the proposedapproach in recognition tasks in the presence of occlusions.