Computers & Graphics 31 (2007) 155–156 Guest editorial Special section on the joint Symposium on Point-based Graphics and Volume Graphics 2006 The drive for increasingly complex 3D geometric models, especially those scanned from the real-world, has brought about a growing interest in methods that build upon point primitives, which is the central focus of Symposium on Point-Based Graphics 2006 (PBG 2006). The Symposium on Volume Graphics 2006 (VG06) is concerned with modeling, processing, analyzing, and rendering of data which are typically acquired through analytical methods, medical scanners, computational simulations, or statistical measurements. Both conferences have been held jointly in the past three years. This special section of Computers and Graphics contains three extended versions of papers presented at the third PBG 2006 and two extended versions of papers presented at the fifth VG06. The papers co-chairs selected these papers by taking into account the detailed reviews from program committee members and the presentation feed- backs at the conference. Authors were invited to submit a revised and extended version of their manuscripts. Each submitted paper went through another round of peer review process based on Computer & Graphics standards. The first paper of PBG 2006, ‘‘A Fast All Nearest Neighbor Algorithm for Applications Involving Large Point-Clouds’’ by Jagan Sankaranarayanan, Hanan Samet, and Amitabh Varshney addresses one of the most fundamental operations of all point-based algorithms, which is to find k-nearest neighbors of a point. Any improvement on the computation and memory access efficiency of this operation will be of benefit to point-based methods, let alone an order of magnitude improvement that has been achieved in this paper. The natural accommodation of out-of-core execution of the method makes this approach even more attractive for truly large point clouds. High quality point-based rendering has been taking a two-pass approach visibility splatting followed by color splatting and compositing; the extra first pass is necessi- tated due to the lack of explicit connectivity among points. Nevertheless, it would be desirable to remove this extra pass and it is the time. Yanci Zhang and Renato Pajarola propose a method to do just that in their paper ‘‘Deferred Blending: Image Composition for Single-Pass Point Rendering’’. This is empowered by a new concept of deferred blending and a reformulation of point interpola- tion into image compositing, which are then accelerated by GPU. Only when achieving high-quality shading of transparent point surface does the method require an additional rendering pass. While points have been demonstrated effectively in many applications, they by no means will completely replace polygonal meshes. In fact, it will add power to points if polygonal meshes can be effectively and efficiently con- structed from them. But how to construct polygon meshes for large (and ever growing) point clouds remains a challenging issue. In ‘‘A flexible framework for surface reconstruction from large point sets’’ Re´ mi Alle` gre, Raphae¨lle Chaine, and Samir Akkouche present a new method to achieve this in a practical sense. A new data- structure that mixes a KD-tree with the Delaunay triangulation is introduced and is the key to efficiently handle point set models with millions of sample points and facilitates out-of-core development. The first selection of VG06 is concerned with acquisition of volumes from a very non-traditional source. The contribution titled ‘‘Density Estimation for Dynamic Volumes’’ by Christian Fuchs, Tongbo Chen, Michael Goesele, Holger Theisel, and Hans-Peter Seidel describes the use of commodity and off-the-shelf camera systems towards the creation of instantaneous volumetric density of dynamic scattering media. A single image is used to create a complete 3D density volume. While fast rendering methods have been the obsession of several volume graphics researches for many years, it is still a challenge to create environments that aid the comprehen- sion of complex volumetric models. The second selection for VG06, ‘‘Interactive Segmentation of Image Volumes with Live Surface’’ by Christopher J. Armstrong, Brian L. Price, and William A. Barrett, focuses on the intuitive and real-time interaction with volumetric models. The pre- sented work emphasizes on the usability and nimble interaction of the realized software thereby enabling a greater understanding of volumetric data. We hope that you will find these five contributions interesting and useful to your own research endeavor. The ARTICLE IN PRESS www.elsevier.com/locate/cag 0097-8493/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.cag.2007.01.019