Vertex space analysis and its application to model based object recognition

The author discusses some of the properties of vertex space, including insensitivity to changes in scale, orientation, and partial object occlusion, and how these properties relate to problems in model-based object recognition. He also describes techniques developed for 2-D and 3-D object recognition using vertex space. The vertex-space approach to object recognition is powerful and efficient, deals well with missing information, and does not require conventional region segmentation. Results to date and an indication of future research directions are presented