Affine-similar shape retrieval: Application to multiview 3-D object recognition

The feasibility of representing a three-dimensional (3-D) object with a sma ll number of standard views is studied in this paper. The object boundary o f each view is considered as a two-dimensional (2-D) shape and is represent ed by the locations of the maxima of its curvature scale space (CSS) image contours, The idea is to identify an unknown object from an image taken fro m a random view by using the stored descriptions of the standard views. The CSS image has been selected for MPEG-7 standardization. The maxima of C SS image have already been used to represent 2-D shapes in different applic ations under similarity transforms. Since the new application involves affi ne transforms, we first examine the effects of general affine transforms on the representation and show that the locations of the maxima of the CSS im age do not move dramatically even under large affine transformations. Our system for shape-based retrieval from large image databases is then app lied to multiview 3-D object representation and recognition. Our collection of 3-D objects consists of 18 aircrafts of different shapes. Three silhoue tte contours corresponding to random, views are separately used as input fo r each object. Results indicate that robust and efficient 3-D free-form obj ect recognition through multiview representation can be achieved using the CSS representation.