The University of Edinburgh -
Division of Informatics
Forrest Hill & 80 South Bridge

PhD Thesis #9609

Title:Recognizing Three-Dimensional Objects Using Parameterized Volumetric Models
Date: 1996
Abstract:This thesis addresses the problem of recognizing 3-D objects, using shape information extracted from range images, and parameterized volumetric models. The domain of the geometric shapes explored is that of complex curved objects with articulated parts, and a great deal of similarity between some of the parts. These objects are exemplified by animal shapes, however the general characteristics and complexity of these shapes are present in a wide range of other natural and man-made 3-D objects.In model-based object recognition three main issues constrain the design of a complete solution: representation, feature extraction, and interpretation. This thesis develops an integrated approach that addresses these three issues in the context of the above mentioned domain of objects. For representation I propose a composite description using globally deformable superquadrics and a set of volumetric primatives called geons: this description is shown to have representational and discriminative properties suitable for recognition. Feature extraction comprises a segmentation process which develops a method to extract a parts-based description of the objects as assemblies of deformable superquadrics. Discontinuity points detected from the images are linked using an "active contour" minimization technique, and deformable, superquadric models are fitted to the resulting regions afterwards. Interpretation is split into three components: classification of parts, matching, and pose estimation. A Radial Basis Function [RBF] classifier algorithm is presented in order to classify the superquadrics shapes derived from the segmentation into one of twelve geon classes.

[Search These Pages] [DAI Home Page] [Comment]