Abstract: This paper presents a new spectral synthesis method, motivated by the spectral turbulence theory, for visual modeling and the realistic and computer-effective animation of turbulent gaseous motion in 2-D and 3-D, which avoids the computational costs of direct simulation. Animation of gases is achieved through a phase shift in the frequency domain according to Kolmogorov's exponential law. The parameters of the turbulence model are intuitive, enabling animaters and designers to become familiar with them quickly. Owing to the method's computational effectiveness, an interactive, (quasi-)real-time, parallelized version, which provides the fast feedack needed to easily adjust the model's parameters, could be implemented.

Abstract: A method for visualizing implicit surfaces is discussed. This method makes as few assumptions as possible concerning the surface and representation of its defining function. The surface may contain singularities, for instance, because it has self-intersections or it is reducible. A userdefined part of space is filled by a set of cubes, cutting pieces (called facets) off the surface. The set of cubes is controlled by an octree converging to the surface. The set of resulting facets can be taken as a piecewise linear approximation, which is sufficiently close to the given surface with respect to criteria specified by the user. Finally, some examples obtained with this method are presented.

Abstract: A variety of halftone methods for reproducing gray-level images on bilevel display media have been proposed. The output of even the best of these falls far short of the quality achieved on similar resolution media in man-made engravings. We introduceDigiDurer — a digital engraving/halftoning system. For a gray-level image input, the system produces a bilevel (black and white only) picture, which has the appearance of an engraving of the input. The system produces high-quality output, even when the graphic elements of the halftone are visible, and both the quality and style of the output can be improved or customized either by using further information on the image content or by interactive user intervention. The heart ofDigiDurer is a curve evolution algorithm generating halftones by controlling the density of line elements, which are the level contours of a potential field induced by the image via an Eikonal equation. Since the basic version of the system produces rather rough results, further capabilities were added to allow for user intervention and the use of image content information. Extensions to graphic elements other than lines are also feasible.

Abstract: We first present the volume-rendering pipeline and the most typical of the existing methods for each pipeline stage. The complexity of each stage in terms of computing time is analyzed for each method. Then the demands and the scope of interactive volume rendering are briefly summarized. Based on this analysis we examine alternate solutions to optimize each pipeline stage in order to allow interactive visualization while maintaining the image quality. The proposed method maximizes interactive manipulation possibilities and minimizes runtimes by sampling at the Nyquist rate and by flexibly trading off quality for performance at any pipeline level. Our approach is suitable for rendering large, scalar, discrete volume fields such as semitransparent clouds (or X-rays) on the fly.

Abstract: This paper describes the application of self-organizing neural networks on the analysis and visualization of multidimensional data sets. First, a mathematical description of cluster analysis, dimensionality reduction, and topological ordering is given taking these methods as problems of discrete optimization. Then, the Kohonen map is introduced, that orders its neurons according to topological features of the data sets to be trained with. For this reason, it can also be called a topology-preserving feature map.

Abstract: Conceptual design and editing parameterized models of 3D solids often requires user interactions with many types of entities and relations. In this paper, we present a concept for simplifying the design process. The simplification is based on three paradigms; integration of data structures and control structures, multiple view interface, and use of hypertext techniques for fast traversal and selective display of data. Data structures and control structures are integrated into uniform building blocks called cells. Cells define assemblies of CSG solids and can be grouped bottom-up into more complex cells or, alternatively, can be refined top-down during the design process. Efficient specification of complex cells and patterns is made possible through the use of automatic default values, multiple instantiation, built-in repetition, and recursion. A multiple-view interface enables the designer to view and edit both the cell parameters and the hierarchical structure of the model. The interface provides integrated textual, 2D graphical, and 3D direct-manipulation techniques for specifying primitive dimensions and relative transformations.

Abstract: Many branches of the sciences produce sample data not arranged on rectilinear grids, and visualization of such data is generally more complex than visualizing more regular data. While isosurface algorithms such as “marching cubes” can be adapted with relatively little performance penalty, direct volume rendering presents more difficulties. This paper explores these issues, concentrating on algorithms for a particular kind of irregular grid known as a curvilinear grid, composed of six-sided cells that may be considered a result of the deformation of a regular rectilinear grid.

Abstract: We present an algorithm for the halftoning of greyscale image sequences. This facilitates the display of video sequences on black-and-white visual displays (e.g., high-qualityX-terminals) for multimedia applications. The main problem to be overcome when halftoning sequences is the temporal flicker between successive images. The classical problem of halftoning a static greyscale image may be posed as an optimization problem. We present an iterative algorithm for its solution. At the expense of being slower, our algorithm achieves visual results on static images better than those obtained from classic halftoning algorithms. We extend our static halftoning algorithm to image sequence halftoning. Temporal correlation between the halftoned versions of image sequences is guaranteed by using an incremental algorithm transfering information between the images. This results in binary sequences with relatively little high-frequency temporal noise, a feature that facilitates efficient no-loss compression (1∶4) of the results.

Abstract: This work in the area of modeling of natural phenomena in computer graphics presents a new, physically based model for the natural process of icicle growth and implements the model creating photorealistic icicle images at the various stages of their growth. The physical model developed here is based on the analysis of the mechanical and thermodynamical processes involved (surface tension, contact angle and wetting, heat transfer, and thermal conductivity). The model is simulated by discrete simulation, yielding an icicle-surface description. This description is converted to Bezier patches and then raytraced.

Abstract: Many systems support the design of two-dimensional (2D) regions and three-dimensional (3D) volumes by sweeping a generator contour over a spine Well-established and widely used rendering systems accept high-level boundary descriptions of regions and solids but not in the contour-spine form Hence the swept object design specification must be such as to lend itself to efficient evaluation of a compact and high-level boundary description In this paper, we identify a sufficiently general class of swept objects and classify the sweep rules We also present computational methods for directly evaluating the boundary representation which implicitly simulate the sweep process The boundary is generated as a set of piecewise curves/surfaces in a compact form suitable for direct input to rendering systems like Post-Script(R) in 2D and Renderman(R) in 3D

Abstract: Presented is a fast display method for volumetric data sets, which involves a slicebased method for extracting potentially visible voxels to represent visible surfaces. For a given viewing direction, the number of visible voxels can be trimmed further by culling most of the voxels not visible from that direction. The entire 3D array of voxels is also present for invasive operations and direct access to interior structures. This approach has been integrated on a low-cost graphic engine as an interactive system for craniofacial surgical planning that is currently in clinical use.

Abstract: A direct algorithm for computing constrained Delaunay triangulation in 2-D is presented. The algorithm inserts points along the constrained edges (break lines) to maintain the Delaunay criterion. Since many different insertions are possible, the algorithm computes only those that are on the Delaunay circles of each intersected triangle. A shelling procedure is applied to put triangles together in such a way that completeness and correctness are guaranteed.

Abstract: The process visualization system of a manmachine interface is presented. It allows human operator(s) to interact with a dynamic technical system, here a thermal power plant. Also, a graphical editor for designing dynamic pictures of the process visualization system is explained. Both new systems have a common architecture and are based on the computer graphics standard PHIGS-2D with extensions for handling non-graphical and graphical information. The hierarchical information structures are manipulated with lists and pointers in a dynamic visual database. The two menu-driven interfaces for designers and for operators are described with their multi-window display layouts. The software of both systems is highly portable to different hardware and to other application domains.

Abstract: This paper presents a novel parallel-processing method for image synthesis using incremental ray tracing on a shared-memory multiprocessor workstation. The most efficient technique for image synthesis is ray tracing, proposed by Whitted in 1980. Ray-tracing algorithms are simple and can generate realistic images. However, they are time-consuming, since calculations of the intersections between objects and ray increase exponentially as the complexity of scenes increases. Fast image synthesis for animation is one of the most important topics in computer graphics. As the area of computer applications has broadened, the complexity of images to be synthesized has increased. Parallel processing of computer graphics is one way of achieving fast image synthesis. This paper describes a parallel processing technique for incremental ray tracing, which recalculates only the rays changed by moving objects in successive scenes of continuous image synthesis. The performance of parallel ray tracing was evaluated on the multiprocessor workstation TOP-1. Strategies for allocating pixels to processes under a multiprocess operating system on this workstation are discussed.

Abstract: Scientific data are often sampled at unstructured spatial locations because of physical constraints, yet most visualization software applies only to gridded or regular data. We discuss several effective techniques for representing scalar and vector-valued functions that interpolate to irregularly located data. Special attention is given to the situations in which the sampling domain is a 2D plane, a 3D volume, or a closed 3D surface. The interpolants can be evaluated on a fine regular grid and they can then be visualized with conventional techniques. Instead of giving a comprehensive survey of this subject, we concentrate on several methods that were developed in the last couple of years.

Abstract: The visualisation of vector fields usually proceeds via geometric forms, such as arrows or streamlines. This paper describes an effective method for volume rendering vector fields in which vectors are assigned colour using a perceptually based colour map. The method is capable of displaying vector fields sampled at voxel rates in three dimensions. This is in contrast to geometric methods that require much lower sampling densities. The approach described automatically colour codes the local field attributes, such as critical points, and enables their emphasis via data windowing techniques. Case studies using gravitational fields and computational fluid dynamics demonstrate the method.

Abstract: The increasing availability of parallel computing platforms has led to the development of parallel volume-rendering algorithms. In the present paper we compare two algorithms for volume raytracing in a data-parallel framework: a shearing technique and a line-drawing technique. The two algorithms are primarily distinguished by the level of parallelism they exploit. Both algorithms have been implemented on the Connection Machine CM2 massively parallel computer and execute at speeds suitable for interactive volume-rendering applications. Since considerable floating-point resources are available on the CM2, we have used rendering algorithms based on transport theory. In the second part of the paper we examine some of the tradeoffs involved between image quality and rendering speed when using high-fidelity rendering algorithms.

Abstract: This paper is concerned with free-form surface constructions using implicit quadrics. More specifically, we are interested in the following problem: given a polyhedron with triangular facets and tangent planes prescribed at its vertices, fit a smooth (tangent-plane continuous), implicit, and piecewise quadric surface through the vertices of the polyhedron so that the surface is tangent to the prescribed tangent plane at each vertex. We show that in order to solve this problem without splitting the facets of the polyhedron, the prescribed tangent planes must satisfy a condition, and under this condition we give a local scheme for constructing the smooth piecewise quadric surface. Using this scheme, we can represent arbitrary shapes by quadric primitives. The implementation results are reported.

Abstract: This paper describes the design and implementation of Mistral-3, a parallel solid odeller for distributed-memory multicomputers implementing algorithms for creating, ray tracing and incrementally updating octree-based spatial data structures. The system has been designed to be scalable and portable. Scalability is addressed by distributing the control of task and data management between all available processors and portability's solved by building software for the abstraction of a distributed-memory multicomputer. An object-oriented approach to the design is adopted to help produce a structured system. Results are presented showing Mistral-3 has made good use of up to 128 T800 transputers. It has been ported to transputer-and non-transputer-based multicomputers.

Abstract: Most two-pass rendering methods calculate a radiosity shading for each patch or element in a scene in the first pass. This shading contains two components: one for the light received directly from the main light sources and one representing the intensity of the light received indirectly by means of diffuse and specular interreflection between patches. However, it is very difficult to achieve accurate representation of the distribution of this radiosity shading over the patch, particularly where clearly visible shadow boundaries exist. A better approach is to store only the indirect reflection component in the form of radiosity shading, and to calculate the direct reflection component during the second pass by casting shadow rays. This approach normally requires that many shadow rays must be cast. However, the number of rays for shadow testing can be kept low by selecting only those light sources that substantially contribute to the shading of a patch and applying an adaptive image refinement technique in combination with a shadow coherence method.

Abstract: A simple recursive procedure that grows integer matrices with a fractal nature is presented. Colorful translations of subsets of these matrices produce aesthetically appealing patterns that extend the, known classes of fractal images. This is illustrated with several examples.

Abstract: Networks based on the High Performance Parallel Interface (HIPPI) will become the norm at Los Alamos National Laboratory. The ramification of such a high-speed networking paradigm on scientific visualization are enormous. Not only will scientists have the capability of networked framebuffer animation loops in their offices, but also the partitioning of graphics tasks between MIMD, SIMD, and specialized hardware will be feasible. Of course, as bandwidth increases, the problem size quickly grows to exceed whatever the limits are. For this reason, the investigation of gigabyte networks is currently underway at Los Alamos National Laboratory.

Abstract: In the context ofimage communication and interchange, to some extent, it comes to philosophical categories, of one tries to explain the term “image”:

Abstract: Ray tracing has been shown to be an excellent technique for rendering realistic images. However, it is important to reduce the lengthy computation time resulting from the brute-force nature of the original ray-tracing algorithms. In this paper, two ideas are proposed to speep up the well-known hierarchical subdivision method. First, a new hierarchy traversal scheme is described to reduce the number of raybounding volume intersection tests for secondary rays. Then, a plane-sweep method is proposed to make fewer intersection checks for eye rays. Experiments and discussions are presented to prove the feasibility of our methods.

Abstract: This paper describes the integration of computer-supported cooperative work (CSCW), multimedia, and hyperstructures into a single framework calledCooperative Hyper Media (CHM). The concept of CHM supports groups of users acting on a single hypermedia document. Cooperative HyperMedia is a layered concept that integrates several notions related to the production of hypermedia documents, namely, document contents, access, and organization. To test the applicability of the concepts of this approach to CSCW we have realized a generic CHM editing architecture (Co MEdiA), telecommunication mechanism for multimedia objects (Tele Media), and service for cooperative work over long distances for time-dependent media (MISTER COOL).

Abstract: Physics-based animation can be generated by performing a complete dynamical simulation of multibody systems. This leads to the solving of a complex system of differential equations in which biomechanical results for the physics of impacts are incorporated. Motion control is achieved by interactively modifying the internal torques. Realtime response requires the distribution of the workload of the computation between a high-speed compute server and the graphics workstation by means of a remote-procedure call mechanism.

Abstract: When displaying abstract objects, as in the case of scientific visualization, it is manadatory to give the user a maximum of cues to help him understand the displayed structures. In this short note, we describe two general, simple 2D postprocessing tools and their applications for the improvement of 3D perception by means of simulation of realistic 3D effects.

Abstract: The 4×4 determinant method makes it possible to unify geometric processing by the computations of 4×4 determinants composed of homogeneous coordinants vectors of four points or coefficient vectors of four plane equations. Because the method needs not require a division operation, error-free geometric computation is not difficult to realize by means of integer arithemtic of appropriate data length. The present paper proposes an error-free, efficient computing method, which computes the 4×4 determinants by adaptively selecting integer arithmetic of variable data length. This technique is discussed theoretically and experimentally.

Abstract: Cooperative hypermedia means producting and manipulating hyperorganized multimedia data by a group of (co-)users. We have been realizing a prototype that enables coauthors to cooperatively produce hypermedia documents. It allows coauthors to communicate their ideas, drafts, guidelines, and constraints within a group in order to exchange information (remotely or face-to-face) and improve the final document.