Contour-Edge Based Polyhedron Visualization.
Lutz Kettner · 2000
Abstract We present a new approach to computing visibility for threedimensional polyhedral surfaces based on contour edges. The advantages are supported by our extensive experimental study of contour edges [11]. We describe preprocessing possibilities for contour edges, a fast silhouette computation algorithm, a new object-space hidden-surface removal algorithm for three-dimensional polyhedral surfaces, and its specialization for terrains. 1 Introduction Given a polyhedron in three-dimensional space and a view point, an edge of the polyhedron is called contour edge if one of the two incident facets is directed towards the view point, and the other incident facet is directed away from the view point (Figure 1). Algorithms on polyhedral surfaces can exploit the fact that the number of contour edges is usually much smaller than the number of edges. In a classification of ten hidden-surface removal algorithms Sutherland introduced in 1974 the distinction between imagespace and object-space algorithms [15]. Object-space methods are characterized to compute the exact answer to the hidden-surface removal problem, which link them to techniques used for visibility problems addressed in computational geometry. A recent survey by Dorward gives an extensive overview of object-space hidden-surface removal algorithms [6]. The result of an object-space hidden-surface removal algorithm is an exact description of the visible parts of a three-dimensional polyhedral surface. A common representation is the visibility map containing the projections of all visible parts including their incidence relations.