Welcome to the special section on volume graphics and point-based graphics. Volume graphics deals with the analysis, synthesis, and presentation of volumetric phenomena, both static and time varying. This includes topics related to the acquisition, reconstruction, and transformation of volume data as well as feature analysis, information extraction, and rendering. Point-based graphics deals with the use of point-based primitives in modeling, rendering, data acquisition, simulation, and geometry. Both volume and point-based graphics have become increasingly important in many applications.
The papers published in this special section are extended versions of papers presented at the IEEE/EG International Symposium on Volume Graphics, September 2007 in Prague, Czech Republic and the joint event of the IEEE/EG International Symposia on Volume Graphics (VG '08) and Point-Based Graphics (PBG '08), August 2008, in Los Angeles, CA (colocated with ACM SIGGRAPH 2008). The VG conferences were held under the auspices of IEEE VGTC and Eurographics, in cooperation with ACM SIGGRAPH. VG '08 and PGB '08 were the seventh and fifth event, respectively, in two series of international symposia. The editors of this special section acted as paper chairs for the last two conferences on volume graphics.
The submissions for both conferences were peer-reviewed. Supported by best paper committees, the paper chairs selected three papers from each event and invited the authors to publish extended versions of these papers in the IEEE Transactions on Visualization and Computer Graphics ( TVCG). The extended versions underwent the standard review procedure of TVCG.
In the following, we summarize the articles, starting with three papers that arose from VG 2007 and three from VG 2008 and PBG 2008:
"Subdivision Analysis of the Trilinear Interpolant" by Hamish Carr and Nelson Max: Isosurfaces are fundamental volumetric visualization tools. For data on cubic grids, usually trilinear interpolation is chosen. An extraction algorithm that is consistent with the trilinear interpolant was presented by G.M. Nielson (2003). The formal proof that these cases are the only ones possible and that they are topologically correct is difficult to follow. This paper presents a more straightforward proof of the correctness and completeness of these cases based on a variation of the Dividing Cubes algorithm. Since this proof is based on topological arguments and a divide-and-conquer approach, it also sets the stage for developing tessellation cases for higher order interpolants and the quadrilinear interpolant in four dimensions.
"Local Ambient Occlusion in Direct Volume Rendering" by Patric Ljung, Frida Hernell, and Anders Ynnerman: In this paper, an efficient method for inclusion of local ambient and emissive lighting for volume rendering applications is presented. Restricting the ambient occlusion to a local neighborhood, combined with a multiresolution framework and adaptive sampling, enables rendering at interactive speeds. A delay is only introduced by the recalculation of the occlusion volume when transfer functions are changed. Uninterrupted user interaction is enabled by progressive refinement. Since the proposed method is gradient-free, it is especially beneficial for noisy data sets. The method gives a better understanding of the shape and density of tissues and so has the potential to increase the diagnostic value of medical volume rendering.
"Per-Pixel Opacity Modulation for Feature Enhancement in Volume Rendering" by Stéphane Marchesin, Jean-Michel Dischler, and Catherine Mongenet: A volume rendering technique is presented that instead of using an opacity transfer function is employing a "relevance function," based on the relative importance of each voxel. This function is used to adjust the opacity of the contributions per pixel. The authors present experiments with different relevance functions and show that this method is suitable for volume rendering at low extra costs. It avoids feature visibility limitations, e.g., by large opaque areas in the data set, and yet maintains a visual similarity with standard volume rendering.
"Illustrative Volume Visualization Using GPU-Based Particle Systems" by Roy van Pelt, Anna Vilanova, and Huub van de Wetering: Illustrative techniques are generally applied to produce stylized renderings that convey visual information effectively. The authors adopt particle systems to produce user-configurable stylized renderings from the volume data, imitating traditional pen-and-ink drawings. In a GPU-based framework, isosurfaces are sampled by evenly distributed particle sets, delineating surface shape by illustrative styles. The appearance of these styles is based on locally measured surface properties. For instance, surface shape is conveyed by orientation of hatches and shape characteristics like curvature are enhanced by color, using curvature-based transfer functions.
"Isodiamond Hierarchies: An Efficient Multiresolution Representation for Isosurfaces and Interval Volumes" by Kenneth Weiss and Leila De Floriani: An interactive particle-based illustrative volume-rendering framework is presented. A novel GPGPU paradigm allows data parallel execution of both the particle system and algorithms, like searching and sorting, in the framework. The flexible particle system can be used for visualization applications, as well as simulation applications, for instance in fluid dynamics, meteorology, and geophysics. Various illustrative styles that resemble pen-and-ink drawings can be applied interactively to isosurfaces in volumetric data. Density and scale-based approaches are used to apply stippling on an isosurface. Similarly direction and curvature-based hatching approaches are presented.
"Fast Construction of k-Nearest Neighbor Graphs for Point Clouds" by Michael Connor and Piyush Kumar: In many applications, including computer graphics, visualization, pattern recognition, computational geometry, and geographic information systems, the problem of computing k-nearest neighbor graphs occurs. In graphics and visualization, this forms a basic building block in solving problems including normal estimation, surface simplification, finite element modeling, shape modeling, watermarking, virtual walkthroughs, and surface reconstruction. This paper presents the design and implementation of a simple, fine-grain parallel and cache-efficient algorithm that uses Morton ordering to solve the k-nearest neighbor graph computation problem for cache-coherent shared memory multiprocessors.
We thank all of the authors for their contributions, their enthusiasm, and their energy. We are particularly grateful to the referees for their detailed reviews and insightful comments, which strongly supported the evolution of the papers. We also thank the editorial board of TVCG for their support and patience.
David H. Laidlaw
H.-C. Hege is with Zuse Institute Berlin (ZIB), Takustr. 7, 14195 Berlin-Dahlem, Germany. E-mail: email@example.com.
D.H. Laidlaw is with the Computer Science Department, Brown University, Box 1910, Providence, RI 02912. E-mail: firstname.lastname@example.org.
R. Machiraju is with the Department of Computer Science and Engineering, The Ohio State University, Columbus, OH 43210.
For information on obtaining reprints of this article, please send e-mail to: email@example.com.
is head of the Visualization and Data Analysis Department at Zuse Institute Berlin (ZIB). After studying physics and mathematics, he performed research in computational physics and quantum field theory at Freie Universität Berlin (1984-1989). Then, he joined ZIB, initially as a scientific consultant for high-performance computing and then as head of the Scientific Visualization Department, which he started in 1991. His group performs research in visual data analysis and develops visualization software such as Amira and Biosphere3D. He is also cofounder of Mental Images (1986), Indeed-Visual Concepts (1999) (now Visage Imaging), and Lenné3D (2005). He taught as a guest professor at Universitat Pompeu Fabra, Barcelona, and as a honorary professor at the German Film School (University for Digital Media Production). His research interests include visual computing and applications in natural sciences, life sciences, and engineering.
David H. Laidlaw
received the PhD degree from Caltech in computer science, where his research centered around how to extract geometric geometric information from volumetric magnetic resonance imaging data and how to optimally acquire such data. He then did three years of postdoctoral research in the Caltech Division of Biology applying image and acquisition results to help advance research in developmental neurobiology. He is a professor of computer science at Brown University. Dr. Laidlaw has published more than 70 peer-reviewed journal and conference papers, has served on or cochaired dozens of conference committees, and has been an associate editor of the IEEE Transactions on Visualization and Computer Graphics
). He has been a recipient of several best-poster, best-case-study, and best-panel awards from IEEE Visualization, two best-student-poster awards from ACM SIGGRAPH, placed first with a collaborative submission to the 2008 US National Science Foundation (NSF)/Science International Science and Engineering Visualization Challenge, and received the 2008 IEEE/VGTC Visualization Technical Achievement Award. He is a senior member of the IEEE.
serves on the faculty of computer science and engineering at The Ohio State University. He has also received appointments at the Comprehensive Cancer Center and the Department of Radiology. His main interests include developing visualization techniques in the biological sciences.