Computational Visualization

Practically all images which meet the human eye today can, in principle, be classified as computer graphics. Some images are produced from raw numeric, symbolic or geometric data obtained from application programs. Most diagrams are produced with spreadsheets or drawing programs. Many photographs are scanned and postprocessed to improve the quality of the colors; such graphics are tuned in a highly interactive process involving users who are experts at such tasks.

Freeform surfaces are very common in the design and modeling of objects, especially when modeling complex objects with curved surfaces. The advantage of working with freeform surfaces is the ability to control the shape of a surface with control points so that the location of the control point influences a well defined part of the curved surface. The greater the complexity of the curve, the greater the number of control points. For a deeper insight into the mathematical matter of freeform surfaces we refer to [Far91, Sei93].

In Chaps. 4 and 5, several analytical approaches for creating line drawings were introduced. These methods allow the generation of a broad variety of drawing styles, and some applications of scientific and medical illustrations were given.

The advantages of 3D computer graphics are being exploited in an increasing number of interactive applications. This not only includes classical fields of 3D graphics (such as CAD or geometric modeling) but also fields in which static 2D images have been used so far (e.g., interactive illustration systems and online catalogs).

Research on reducing the interaction effort when navigating in large information spaces has utilized Furnas’ generalized fisheye view [Fur86] in a variety of techniques (cf. Chap. 2). Existing approaches apply filtering and distorting techniques or both (recall Noik [Noi94]) according to the degree of interest (DOI) at a certain point in the information space (node). “Classical” fisheye views, as described by Sarkar and Brown [SB94], distort an existing layout to enlarge the view at the point of interest (focus). Distorting fisheye views stick more to the photographic nature of fisheye lenses by applying a nonlinear distortion to the display transformation. Filtering fisheye techniques display or suppress the rendering of nodes according to their DOI. Hybrid techniques, like the Intelligent Zoom as described by Bartram [BOD⁺94], make use of both, applying fisheye distortion to the display transformation as well as choosing an appropriate representation for each node not only depending on its DOI value, but also using reasoning techniques to exploit additional contextual information. Semantic zooming as described in [PF93] changes the appearance of objects, too, although only depending on the level of detail (LOD).

This chapter will concentrate on the relationship between images and text, and in particular on the use of text to present nominally graphical information to blind people. The following chapter will then consider the role of text within images. First, we look at the role of text, both as a component within graphical images, as a complement to graphical images, and as an alternative means of conveying information.

Visualizations are produced to enable a viewer to extract information. For this purpose, visualizations are not merely a straightforward rendering of the data. Limited presentation space imposes restrictions on the visualization, resulting in omissions, exaggerations, or displacements. Moreover, viewers may wish to explore the data under a thematic focus. Thus, portions of the data may be presented in more detail or more comprehensively, while others may be simplified, shrunken, or even left out.

Interactive 3D graphics have a high potential for the explanation of complex spatial phenomena as can be found, for example, in engineering and anatomy. The interactive exploration of complex 3D models is crucial for spatial understanding. While this is well recognized, not enough effort has been spent on flexibly combining interactive 3D graphics with textual descriptions.

This chapter deals with the creation of 3D non-photorealistic computer animations. It reviews the basic concepts for traditional and computer-generated animation with an emphasis on non-photorealistic animation which, when examined closely, covers more than just the mere rendering of a number of subsequent images. This chapter will not discuss new techniques for object deformation or motion specification but concentrates on the visualization of moving non-photorealistic images. As we will see, this requires new rendering methods for “drawing” a picture.

This chapter is focused on extending illustration techniques to sequences of frames — animations. Computer animations offer additional degrees of freedom to emphasize objects, to show relations between objects, and to clarify their spatial arrangement by performing gradual frame-to-frame changes in the rendered images. Object transformations, changes of an object’s material, and control of a virtual camera provide interesting channels to convey information.

Many authors have observed the impact of the increased cognitive load users are confronted with when using today’s high-functionality computer systems. Graphical (direct manipulation) interfaces tend to use the notion of infinite working sheets. Yet existing screen real-estate is finite, partly due to current technological restrictions but mainly because of the limitations imposed by the human perceptual system.

In anatomy, and even more in surgery, the viewer’s expectation of a good illustration is to get reliable information about form, structure, and sometimes also texture of the depicted objects. Within printed illustrations this can be achieved by using different kinds of drawings or even photographs. Here structure can be depicted by color coding, texture by “shading” the surface differently. The general presentation goals, as stated in Chap. 4, also apply here. Besides that, medical illustrations include an artistic handling of the subject and thus a medical illustration can be regarded as a fusion of science and the graphic image.

In this chapter, computer generated illustrations and animation sequences of hand gestures are discussed. The animation of gestures is very useful in the teaching of sign language.

While the previous chapter concentrated on how to render animations, this chapter deals with how to create them. It introduces a new method of generating the animation content mentioned in Chap. 15. That content can be presented using the techniques described in that chapter.

Like maps for sighted people, tactile maps are depictions of areas like countries or towns. However, they are not just embossed versions of the maps that sighted people are used to (see Fig. 20.1). A number of particularities have to be observed to ensure the tactual legibility of these maps, resulting in several restrictions to the map design (refer to [Edm91] and [Hud83] for more information). Since haptic perception has a lower discriminability, symbols have to be larger and simpler, e.g., point symbols should have a diameter of 3–5 mm. To avoid confusion, symbols must have a minimum distance between one another. While these distances are about 0.1mm for visual features, tactile symbols must be at least 2–3 mm apart to ensure their separate perception. Braille texts have a fixed height of 6 mm and hence entail a street width of at least 8 mm to allow labeling of the streets. While color can be replaced by textures, the textured areas have to be very large to enable the differentiation of the different textures, and still there are only very few different textures easily recognizable even for larger areas.

Holography is a method for three-dimensional imaging of objects. It was first presented by Dennis Gabor [Gab48] in 1948 and has expanded into various fields since then. In holographic interferometry, arbitrarily shaped diffusely scattering objects are examined. A description of this process can be found in Lauterborn, Kurz, and Wiesenfeldt [LKW95] and Eichler and Ackermann [EA93]. The same object can be compared with itself at a later time, allowing for the analysis of stress or deformations. Other applications of holography are in the design of diffractive optical elements, as described by Aagedal et al. [ABST94] or holographic storages [EA93].

When browsing through a book on computer graphics, one usually finds a lot of more or less interesting pictures that are produced by means of computers. These pictures are embedded in pages of technical texts describing how this image generation was performed and why it provides a better way to do so than other methods. Less space is usually given to the methodological background and the motivation underlying the preoccupation with computer visualization. In this chapter, we want to complement the more technically oriented part of this book with some reflections as to why such techniques can be interesting not only for computer graphics researchers, and where, from a communication-theoretic point of view, they might be of use in our society.

Our book has dealt with selected aspects of what we call computer visualization. Indeed, we use the term Computational Visualistics¹ for the scientific study of how visualizations are captured, stored, processed, produced, and conveyed to users, as well as how computer users interact with, perceive, understand, and store pictures. The characteristic aspect of computational visualistics is that we always consider algorithms running within the computer in unison with what the user will do with the resultant graphical output. We have dealt with only one specific aspect of the topic of computational visualistics, that of producing and interacting with images of a particular kind. In this chapter, we shall now take a fresh look at the technical results presented, strive for an insight into what is actually happening when producing such images, and work toward placing the results into a wider context. In doing so, we shall also suggest some new concepts and terminology which will guide future work in the area.