Abstract

Our approach to shape modeling is based on parametric description of all aspects of shape: volume, surface, and contour. We advocate this description as an immediate step applied directly to range data. The rationale behind this approach is the following: (1) The representation is compact, i.e., it has a large expressive power with a small set of parameters; (2) the parameters have natural interpretations: position of the object in three-dimensional space, orientation, size, curvature, shape, etc.; (3) the models are recovered from data by using numerical fitting procedures that provide a measure of goodness of fit, which in turn can be used in a feedback loop to improve the description according to the task; (4) the representation provides a predictable segmentation of a scene into its parts.

© 1990 Optical Society of America