ABSTRACT
This chapter discusses the method for the detection of salient nonlocal structures in presentation graphics. MAX simulates the grouping of graphical objects according to the Gestalt grouping principles of proximity, similarity, and good continuation. Arrangement polygons are compared by a polygon matching algorithm that computes a similarity metric for polygonal shapes; this makes it possible to retrieve similarly arranged groups in the Graphics Designer's database of presentation graphics. All found graphics with matching group indices are handled by the search component in its usual way: Some key aspects of Treisman's model will be described in the next section. Treisman's feature maps may be conceived as similarity classes, so we have adapted the preattentive part of the model for our simulation of similarity grouping in MAX. The simulation of proximity grouping is a preparatory step for continuity grouping, because good continuations are formed only between objects that have equal or similar distances.
