FITTING 3D MORPHABLE MODELS USING IMPLICIT REPRESENTATIONS

Curzio Basso and Alessandro Verri

We consider the problem of approximating the 3D scan of a real object through an affine combination of examples. Common approaches depends either on the explicit estimation of point-to-point correspondences or on 2-dimensional projections of the target; both present drawbacks. We follow an approach similar to (Ilic and Fua, 2003) by representing the target mesh via an implicit function, whose values at the vertices of the approximation are used to define a robust cost function. The problem is approached in two steps, by approximating first a coarse implicit representation of the whole target, and then finer, local ones; the local approximations are then merged together with a Poisson-based method. We report the results of applying our method on a subset of 3D scans from the Face Recognition Grand Challenge v.1.0.

SimpliPoly: Curvature-based Polygonal Curve Simplification

Sumanta Guha, Paul Janecek and Duc Cong Song Nguyen

A curvature-based algorithm to simplify a polygonal curve is described, together with its implementation. The so-called SimpliPoly algorithm uses Bezier curves to approximate pieces of the input curve, and assign curvature estimates to vertices of the input polyline from curvature values computed for the Bezier approximations. The implementation of SimpliPoly is interactive and available freely on-line. Empirical comparisons indicate that SimpliPoly performs as well as the widely-used Douglas-Peucker algorithm in most situations, and significantly better, because it is curvature-driven, in applications where it is necessary to preserve local features of the curve.

The heat equation and the Frenet formulas for digital curves

Sheng-Gwo Chen, Jyh-Yang Wu, Mei-Hsiu Chi and Chen-Yao Lai

In this note we shall discuss the heat equation and the eigenvalue problem for digital curves in the 3D Euclidean space. First, we shall introduce the derivative of a function along a digital curve by the weighted combination method. Then, we can define the Laplace of a function on a digital curve. The Frenet formulas for digital curves will also be discussed. Numerical simulations show our method will provide good estimations for the curvature and torsion.

A simple and fast hardware-accelerated point-in-polygon test

Francisco Martínez, Antonio Rueda and Francisco R. Feito

The new generations of GPUs bring us a set of new basic operations or tools that allow us to design new solutions to traditional problems in Computer Graphics. In this paper we present two approaches for the point-in-polygon problem based on the occlusion query extensions supported by modern GPUs. Both approaches are fast and their execution times do not depend on the number of edges of the polygon. Besides, one of the tests allows multiple point-in-polygon queries to be done in parallel with CPU computations.

A TEXTURE-BASED METRIC EXTENSION FOR SIMPLIFICATION METHODS

Carlos González, Pascual Castelló and Miguel Chover

We present an extension of the error metrics used in the simplification methods based on edge collapse operations, where we take into account texture information. Many simplification methods are only based on the geometry of the models, without considering texture information. So, the simplified models present high distorted textures. The presented metric avoids the early collapse of edges that crashes with no uniform regions of the texture. The detection of these regions is performed by an edge detector method based on Canny. For testing the new error metric, an own geometric simplification method based on edge collapses has been used. It can be observed that simplified models with this new metric present more realistic results than before. This metric produces a modification in the order of the edge collapses. It is very useful for multiresolution models. The computational cost of this metric is negligible compared with the simplification time.

CONCAVE HULL: A K-NEAREST NEIGHBOURS APPROACH FOR THE COMPUTATION OF THE REGION OCCUPIED BY A SET OF POINTS

Adriano Moreira and Maribel Yasmina Santos

This paper describes an algorithm to compute the envelope of a set of points in a plane, which generates convex or non-convex hulls that represent the area occupied by the given points. The proposed algorithm is based on a k-nearest neighbours approach, where the value of k, the only algorithm parameter, is used to control the “smoothness” of the final solution. The obtained results show that this algorithm is able to deal with arbitrary sets of points, and that the time to compute the polygons increases approximately linearly with the number of points.

2D IMAGES CALIBRATION TO FACIAL FEATURES EXTRACTION

Daniel Trevisan Bravo and Sérgio Roberto Matiello Pellegrino

The extraction of facial characteristics is too much important for several applications that involve the 3D reconstruction and facial recognition. In general, the facial modeling applications that use 2D images for the 3D reconstruction, it is necessary to prepare the images for the facial features will be extracted in an efficient way. In this paper there are procedures for calibration and correction of two facial images (the frontal and the lateral) acquired in distinct instants. As this study aim is to work with human faces, it is believed that it has well known characteristics, such as the eyes position, mouth, nose curvature, etc. Taking this previous knowledge into account, it was tried to insert it in a general context to make the automatic recognition of the searched parts, building algorithms that identify those characteristics on specific regions of the face. Although this approach makes the process faster, it also imposes restrictions to the model that, however, do not disable its execution in most of cases.

OUT OF CORE CONSTRUCTION OF PATCH TREES

Hermann Birkholz

Current Level of Detail (LoD) approaches for triangle meshes use variably triangulated mesh patches, in order to approximate the original mesh surface. The approximation is synthesized from some of these patches, which must cover the whole surface and must not intersect each other. The patches are chosen ac-cording to the necessary view dependent triangulation. This paper addresses the creation of the required hi-erarchical data structures, in order to enable Level of Detail synthesis for very large triangle meshes. Be-cause of the limited amount of internal memory, most of the mesh data reside in external memory during the process. Due to the high access latency of external memory, commonly used algorithms for small meshes are hardly applicable for so called “Out of Core” meshes. Other methods have to be found, which overcome the problems with the external memory.

Volumetric Snapping: watertight triangulation of point clouds

Tim Volodine, Michael S. Floater and Dirk Roose

We propose an algorithm which constructs an interpolating triangular mesh from a closed point cloud of arbitrary genus. The algorithm first constructs an intermediate structure called a Delaunay cover, which forms a barrier between the inside and the outside of the object. This structure is used to build a boolean voxel grid, with cells intersecting the cover colored black and all other cells colored white. The outer surface of the voxel grid is snapped to the point cloud by replacing each exterior surface vertex with the closest point in the point cloud. The snapped mesh is processed such that it is manifold and consists of triangles with good aspect ratio. We show that if a fine voxel grid is used, the snapping yields Delaunay-like triangulation of the original points. High grid resolutions are possible because of the Delaunay cover and a new contouring method, which extracts the outer surface of the grid with $O(n^2)$ worst case space complexity, where $n$ is the number of voxels in one dimension.

Pyramid Filters Based on Bilinear Interpolation

Martin Kraus and Magnus Strengert

The implementation of several pyramid methods on programmable graphics processing units (GPUs) in recent years led to additional research interest in pyramid algorithms for real-time computer graphics. Of particular interest are efficient analysis and synthesis filters based on hardware-supported bilinear texture interpolation because they may be used as building blocks for many GPU-based pyramid methods. In this work, several new and extremely efficient GPU-implementations of pyramid filters are presented for the first time. The discussion employs a new notation, which was developed for the consistent and precise specification of these filters and also allowed us to systematically explore appropriate filter designs. The presented filters cover analysis and synthesis filters, (quasi-)interpolation and approximation, as well as discontinuous, continuous, and smooth filters. Thus, a toolbox of filters and their efficient implementations for a great variety of GPU-based pyramid methods is presented.

AN ENHANCED EDGEBREAKER COMPRESSION ALGORITHM FOR THE CONNECTIVITY OF TRIANGULAR MESHES

Dina Reda Khattab, Yasser Abd El-latif, Saeed Abd El-Wahab and Fahmy Tolba

Compression of digital geometry models is the answer to an industrial demand. Over the last years, many exciting ideas and new theoretical insights have been devoted to finding ways of reducing the amount of storage such models absorb. EdgeBreaker is one of the effective lossless single-rate connectivity compression techniques for triangular meshes. This paper presents an enhanced EdgeBreaker encoding algorithm which solves the problem of non-linearity of EdgeBreaker decoding procedure while reconstructing the mesh triangles in the same order they were traversed during the encoding phase. The new enhancement is based on the same data structure: the corner-table used by EdgeBreaker however, it eliminates some of the computational overhead exhibited by EdgeBreaker compression. This enhanced technique also yields to significantly smaller rates for connectivity compression than EdgeBreaker. It achieves an average compression ratio of 1.8 bit per triangle and 3.57 bit per vertex for the used benchmark 3D models.

TILING THE SPHERE WITH DIAMONDS FOR TEXTURE MAPPING

John Crider

The sphere can be covered by any of an infinite number of tiling sets of equilateral spherical quadrilaterals (diamonds). Five of these tiling sets have practical use for texture mapping application. Points on the sphere can be described by intersections of geodesics, which provide coordinate values in a new coordinate system, defined for each tiling set. Each of the diamonds can be subdivided by a grid of coordinate geodesics to pixel level and so can be directly mapped to and from a texture array. The diamonds can also be subdivided into spherical quadrilaterals that can be approximated by pairs of triangles for fast rendering in a graphics system. Because coordinates in the new system are readily converted to and from Cartesian coordinates, diamonds can be used easily in interactive graphics and ray-tracing applications.

Orthant Neighborhood Graphs - A Decentralized Approach for Proximity Queries in Dynamic Point Sets

Tobias Germer and Thomas Strothotte

This work presents a novel approach for proximity queries in dynamic point sets, a common problem in computer graphics. We introduce the notion of Orthant Neighborhood Graphs, yielding a simple, decentralized spatial data structure based on weak spanners. We present efficient algorithms for dynamic insertions, deletions and movements of points, as well as range searching and other proximity queries. All our algorithms work in the local neighborhood of given points and are therefore independent of the global point set. This makes ONGs scalable to large point sets, where the total number of points does not influence local operations.

Realistic Transmission Model of Rough Surfaces

Huiying Xu and Yinlong Sun

Materials have three basic types: opaque, transparent and translucent. Transparent and translucent objects involve both light reflection and transmission at surfaces. This paper develops a realistic transmission model of rough surfaces using a physically based approach called the statistical ray method. The surface is assumed locally smooth and statistical techniques can be applied to calculate light transmission through a local illumination area. We have obtained an analytical expression for single scattering. The analytical model has been compared to our Monte Carlo simulations as well as to simulations by others, and good agreements have been achieved. The proposed model has a good application potential for realistic rendering.

Building 3D Indoor Scenes Topology From 2D Architectural Plans

Sébastien Horna, Guillaume Damiand, Daniel Meneveaux and Yves Bertrand

This paper presents a new method for reconstructing geometry and topology of 3D buildings from 2D architectural plans. A complete topological model expresses incidence and adjacency relations between all the elements. It is necessary for both recovering accurately 2D information and constructing a coherent 3D building. In 2D, several high-level operations are proposed for creating walls, portals, stairs, etc. Semantic information is associated with all volumes for specifying openings, walls, rooms, stairs, facade, etc. The resulting 2D model is extruded for generating a 3D environment, taking the semantic information into account since doors are not processed as walls for instance. Floors are superimposed using volumes corresponding to upper and lower ceilings linked according to stairways.

Extracting Terrain Morphology: A New Algorithm and a Comparative Evaluation

Paola Magillo, Leila De Floriani, Emanuele Danovaro, Laura Papaleo and Maria Vitali

We consider the problem of extracting morphology of a terrain represented as a Triangulated Irregular Network (TIN). We propose a new algorithm and compare it with representative algorithms of the main approaches existing in the literature to this problem.

A FAST INTERPOLATION METHOD TO REPRESENT THE BRDF IN GLOBAL ILLUMINATION

Pierre Chatelier and Rémy Malgouyres

Global illumination that simulates realistic lighting environments makes use of bidirectional reflectance dis- tribution functions (BRDF). Such a function is a surface property, describing how light is re-emitted after hitting this surface. The present paper details a representation of BRDF and radiance distributions, based on control points, that aims at performance. It uses spherical triangles for interpolation, least square mean or linear programming for optimal representation, and very fast rotation by precomputing control points re- weightening. It has some more advantages for deforming an existing shape. This approach is compared to Spherical Harmonics, and outperforms this last one.

Morphology-Based Representations of discrete Scalar Fields

Mohammed Mostefa Mesmoudi and Leila De Floriani

Forman introduced in(Forman,1998) a theory for cell complexes that is a discrete alternative to the well known Morse theory. Forman theory is currently finding several applications in digital geometry and image processing since the handled data are discrete, see for instance (Lewiner et al.,2002a) and (Lewiner et al.,2002b). In (de Floriani et al., 2002b) we have introduced a Smale-like decomposition of a scalar field defined on a triangulated domain based on a discrete gradient field that simulates well the behaviour of gradient field in the differential case. In this paper we extend our discrete gradient vector field so that the extended form corresponds to a Forman gradient field. Hence, the extended gradient field inherits properties of both smooth Morse and discrete Forman functions.

Advanced direct manipulation of feature models

Rafael Bidarra, Alex Noort, Pedro Oliveira and Daniel Lourenço

In current commercial feature modelling systems, support for direct manipulation of features is not commonly available. As a result, re-designing is time-consuming due to the inefficient feedback, the insight given is rather poor, and user interaction often lacks intuitiveness. This is partly due to the lack of speed of current constraint solvers, but also to deficient interactive facilities. In this paper, we argue that providing advanced direct manipulation facilities for feature models is possible and can significantly speed up the product design process, by giving designers a much more intuitive interface, with immediate feedback and deeper insight into the consequences of each modelling action. An approach to such a direct manipulation interface is presented that brings together the advantages of direct manipulation of feature models with the necessary emphasis on fundamental feature modelling paradigms like feature parametrisation and feature validity maintenance. In particular, it offers a powerful combination of various 3D handles for real-valued feature parameters, with a preview overlay facility for all modelling operations. Details are provided on how this approach was successfully implemented in a prototype feature modelling system.

LOCAL MULTIRESOLUTION ANALYSIS OF A MESH

Olivier Guillot and Jean-Paul Gourret

We build a local multiresolution analysis of surfaces when the mesh have the topology of the result from the square root 3 subdivision of a coarse mesh template. We use the concept of biorthogonality and lifting to develop a set of filters for local analysis and local synthesis. The method presented here takes into account discontinuities during the subdivision process.

Detecting Features from Sliced Point Clouds

Ioannis Fudos, Ioannis Kyriazis and Leonidas Palios

We present a new method for extracting the feature primitives of a 3D object directly from the point cloud of its surface scan. The objective is to identify a subset of points that provides the same information about the structure and the topology of the object geometry as the point cloud itself. The entire process is carried out with the least human intervention possible, and the only information we receive as input is the point cloud of the 3D scan of the object. First, the point cloud is sliced interactively in cross sections. Each cross section consists of a 2D point cloud. A collection of curve patches is derived for each slice, describing the cross section, and providing the feature based model that we need. For the extraction of the feature points and the interpolating curve patches we use properties of the convex hull and the voronoi diagram of the point cloud. The slices are then assembled together with advanced solid modelling operations to derive a CAD model for the initial object.

A Virtual Environment for Archiving Micro-Presence with Image-Based Model Acquisition

Masahiko MORITA, Tastuya SAITO and Kenji KOHIYAMA

Micro Archiving is a new method of photography that helps anybody digitally archive minute structures that existing capturing systems cannot deal with. For research use, it is an effective way to make high-fidelity virtual specimens of small objects that allow scholars to share the precious academic resources over the Internet in a digital format. For commercial use, the virtual models captured by the technology can be applied to computer graphics, games, digital cinemas and other digital entertainment media. Technically, our Micro Archiving technology enables easy and fast construction of high-quality and high-resolution textured virtual 3D models of small objects. Currently, application contents of Virtual Reality, Augmented Reality and other interactive media are dependent on the use of elaborate 3D models, which require expensive manual task. Micro Archiving technology provides automatic capturing process that can comprehensively handle a high-resolution all-focused imaging, 3-dimensional modeling and transparency capturing simultaneously. Micro Archiving is best suited for minute and complex objects such as machine parts, insects, plants and small archaeological materials.

Hierarchical Brain Model for Coregistration

Terrence Oakes

A ubiquitous problem in coregistration of brain images is that individual sulci and gyri vary considerably between individuals, both with respect to location and shape as well as for simple existence of particular sulci. The underlying assumption of most coregistration processes is that one structure can be smoothly morphed to exactly resemble another structure if enough parameters are used. Although in a strict sense this may be true for intersubject brain registration, due to differing structures the result may not be as meaningful as desired. The proposed approach offers a groundbreaking alternative to the standard approach of continuously deformable coregistration algorithms, introducing instead a hierarchical structure of related nodes (a "nodetree") to model the brain structure using gray-matter and white-matter masks. Additonally, a proposal is made for using the nodetree structure for coregistration, employing a novel locally discontinuous but focused registration to more accurately align and compare corresponding features. This approach can provide a framework for identifying structural differences, with a goal of relating them to functional differences. Although this method uses the brain as an example, it is quite general and not limited to the brain, or even to medical images.

Modeling Dendritic Forms Using Path Planning

David Mould and Ling Xu

We present a method for creating geometric models of dendritic forms. Dendritic shapes are commonplace in the natural world; some examples of objects exhibiting dendritic shape include lichens, coral, trees, lightning, rivers, crystals, and venation patterns. Our method first generates a regular lattice with randomly weighted edges, then finds least-cost paths through the lattice. Multiple paths from a single starting location (or generator) are connected into a single dendritic shape. Alternatively, path costs can be used to segment volumes into irregular shapes. The pathfinding process is inexpensive, and admits control handles including endpoint placement, distribution of generators, and arrangement of nodes in the graph.

SURFACE MODELING OF MULTI-POINT, MULTI-FLUTE CUTTING TOOLS

Phalguni Gupta, Sanjay G. Dhande and Puneet Tandon

Cutting tools are usually represented by two-dimensional representation schema(s). The two-dimensional nomenclatures have their inherent limitations. This paper outlines a detailed surface based modeling paradigm for a variety of multi-point, multi-flute cutting tools. The work presents the generic biparametric surface patch based models of slab mills, end mills and drills. A slab mill consists of flutes (meant for cutting) and planar end surfaces, while the end mills and drills are made up of flutes, shanks and varied end geometries. The flutes are modeled as helicoidal surfaces which are generated by first developing the sectional geometry of its tip-to-tip profile and then sweeping according to the type of the cutter under consideration. The transitional surfaces of these cutters are modeled as bicubic Bézier surfaces or biparametric sweep surfaces. The relations to map proposed three-dimensional rotational angles that help define three-dimensional geometric models to conventional angles (forward mapping) and their reverse relations (inverse mapping) are also developed. The new paradigm offers immense technological advantages in terms of numerous downstream applications.

HIERARCHIC MODELING OF SYMMETRIC GEOMETRIES

Pritee Khanna, Puneet Tandon and Sanjay G. Dhande

The world is full of procedural ornaments, especially in cultural heritage. These ornaments are often said to be too complex for automatic modeling, and too tedious for manual modeling. A pattern is an orderly arrangement of objects in space. In this work a geometric model for symmetric patterns is proposed. The two-dimensional hierarchic model represents pattern as a tree. The model when implemented reads in a motif description, builds a tree of motifs and renders the pattern as well as final layout. Interactive editing of motif descriptions can be performed with a GUI. The proposed geometric model for two-dimensional patterns can be incorporated into design software that can be used by persons having minimal artistic skills.

Real-time Rendering of Time-Varying Volume Data Using a Single COTS Computer

Fumihiko Ino, Daisuke Nagayasu and Kenichi Hagihara

This paper presents performance results of an out-of-core renderer, aiming at investigating the possibility of real-time rendering of time-varying scalar volume data using a single commercial off-the-shelf (COTS) computer. Our renderer is accelerated using software techniques such as data compression methods and thread-based pipeline mechanisms. These techniques are efficiently implemented on a COTS computer that combines multiple GPUs, CPUs, and storage devices using scalable link interface (SLI), multi-core, and redundant arrays of inexpensive disks (RAID) technologies, respectively. We find that the COTS-based out-of-core renderer achieves a video rate of 35 frames per second (fps) for 256x256x208 voxel data with 99 time steps. It also demonstrates an almost interactive rate of 4 fps for 512x512x295 voxel data with 411 time steps.

Memory-Efficient View-Dependant Level of Detail of High-Detailed Meshes

Mathias Holst and Heidrun Schumann

In this paper we propose an effective and efficient out-of-core LOD framework for arbitrary irregular high-detailed triangular meshes and scenes. We modify and extend the standard Multi-Triangulation framework to reduce CPU load and to increase RAM efficiency by paging. To reduce the number of IO-operations we propose a novel partitioning scheme well suitable for view-dependant LOD. In addition a fast but very effective caching strategy is proposed, which bases on a LOD prediction for future frames. The efficiency of our framework is shown by results.

Direct Texture Synthesis of Feather Pigmentation Patterns

Marcelo Walter and Cristiano Franco

Feathers present exquisite shapes and visual patterns. Just recently have feathers been addressed in computer graphics with realistic-looking results. Current solutions, however, make use of texture mapping for adding the visual patterns seen on feathers. In this paper we present an integrated model for direct texture synthesis of feather pigmentation patterns. The model uses B\'ezier curves to model the feather shape and procedural MCLONE patterns for the synthesis of feather pigmentation patterns. We show that the Clonal Mosaic model (MCLONE) is a viable option for direct texture synthesis of the many pigmentation feather patterns found in Nature.

NON-PHOTOREALISTIC RENDERING OF ALGORITHMICALLY GENERATED TREES

Cindy Grimm and Nathan Dudley

This paper presents a novel rendering technique inspired by artistic approaches. Instead of trying to recreate a traditional medium, such as charcoal or watercolor, this approach is a mixture of both photo-realism and abstraction. Artists use a process of abstraction to provide structural information about subjects that do not have clearly defined shapes, such as groups of leaves in a tree. For example, an artist will use a color wash to first approximate a group of leaves. They then add detail on top of parts of this wash to indicate the presence of individual leaves. Similarly, we use an abstract shape that approximates the image of leaves clustered at the end of a branch. To prevent oversimplification, we add photo-realistic detail using a blending process. Inter-frame coherence is achieved both by smoothly interpolating the abstract shapes, and the continuity inherent in the photo-realistically rendered detail.

Supercover Plane Rasterization

Tomislav Petkovic and Sven Loncaric

An analysis of a rasterization algorithm for generating supercover planes in 3D voxel space is presented. The derived algorithm is an extension to the classical 2D line rasterization algorithm. Additional voxels needed to form the supercover 3D plane are identified by rasterizing two additional 2D lines per volume slice. A discussion on how to modify the algorithm to rasterize finite supercover 3D plane segments with arbitrary parameters by using integer arithmetic only is given.

Adaptive cube tessellation for topologically correct isosurfaces

Francisco Velasco, Juan Carlos Torres, Alejandro Leon and Francisco Soler

Three dimensional datasets representing scalar fields are frequently rendered using isosurfaces. For datasets arranged as a cubic lattice, the marching cubes algorithm is the most used isosurface extraction method. However, the marching cubes algorithm produces some ambiguities which have been solved using different approaches that normally implying a more complex process. One simple way to produce an isosurface without ambiguities in the process, is to tessellate the cubes into tetrahedra, using a similar method (marching tetrahedra), to build the isosurface. The main drawback of this method is that the tessellation used does not produce the same isosurface topologies as those generated by improved marching cubes algorithms. We propose a tessellation that produces the same topology for all the cases, moreover the tessellation allows isosurface to evolve continuously when the isovalue is changed smoothly without extra computations because the tessellation process is independent of the isovalue.

Quality-Based Improvement of Quantization for Light Field Compression

Daniel Meneveaux, Raphaël Lerbour, Bruno Mercier and Chaker Larabi

In the last decade, many methods have been proposed for rendering image-based objects. However, the number and the size of the images required are highly memory demanding. Compression is thus necessary. Based on the light field data structure, we propose an improved compression scheme favoring visual appearance and random access. Our method relies on vector quantization for preserving access in constant time. 2D Bounding boxes and masks are used to reduce the number of vectors during quantization. Several light field images are used instead of blocks of 4D samples, so that image similarities be exploited as much as possible. Psychophysical experiments performed in a room designed according to ITU recommendations validate the quality metrics of our method.

Hybrid Ray Tracing - Ray Tracing Using GPU-Accelerated Image-Space Methods

Philippe Robert, Severin Schoepke and Hanspeter Bieri

In recent years, interactive ray tracing has become a reality, although mainly by using clustered workstations and sophisticated acceleration structures. On non-clustered computer architectures this is still not an easy task, especially when rendering animated scenes, even though the computation power of modern workstations is increasing rapidly. In this paper we propose known image-space rendering techniques to be used for accelerating ray tracing. Firstly, we describe a GPU-based visibility preprocessing algorithm to perform interactive ray casting by applying the standard depth testing capability of graphics processing units. This method -- called object intersection buffer (OIB) -- is particularly suitable for ray casting animated scenes, as it completely avoids the necessity of creating and updating any kind of spatial acceleration structures in order to achieve high frame rates. Then we integrate shadow rendering into our ray caster using the shadow mapping technique to avoid computationally expensive shadow rays. Then, we convert our GPU-based ray caster into a hybrid ray tracer by computing reflection and refraction rays on the CPU using a spatial acceleration structure. This allows us to exploit parallel rendering to increase the overall frame rate. Finally, we compare our implementations to each other and analyse their advantages and disadvantages in terms of visual quality and rendering performance.

A Hybrid Backward-Forward Method for Interactive Reflections

Voicu Popescu, Chunhui Mei and Elisha Sacks

Rendering reflections is challenging and current interactive techniques trade quality and accuracy for performance. In ray tracing, the desired image rays are traced back into the scene to produce high quality reflections, but finding which primitives intersect each ray is inefficient. The feed-forward graphics pipeline relies on projection to directly find the pixels covered by a primitive. However, feed-forward reflection rendering requires projecting reflected vertices, a difficult problem that has only been solved in the context of planar or convex reflectors. We propose a two phase hybrid reflection rendering method based on approximating the reflected rays with a set of simple cameras modeled as continuous 3-ray cameras. In the first, "backward", phase, the view volume of each simple camera is intersected with a hierarchical subdivision of the scene to find the geometry it encompasses. In the second, "forward", phase the geometry is projected with the simple camera. Since the shape and topology of reflected triangles is complex, point based rendering is adopted to reconstruct the reflection. The hybrid method is efficient since it combines advantages of backward and forward techniques: there are two orders of magnitude fewer simple cameras than reflected rays, the hierarchical scene subdivision implements fast view volume culling for each of the simple cameras, and the reflection piece corresponding to each simple camera is computed efficiently in feed-forward fashion.

INVERSE RENDERING FOR ARTISTIC PAINTINGS

Shinya Kitaoka, Tsukasa Noma, Yoshifumi Kitamura and Kunio Yamamoto

It is difficult to apply inverse rendering to artistic paintings than photographs of real scenes because (1) shapes and shadings in paintings are physically incorrect due to artistic effects and (2) brush strokes disturb other factors. To overcome this difficulty of non-photorealistic rendering, we make some reasonable assumptions and then factorize the image into factors of shape, (color- and texture-independent) shading, object texture, and brush stroke texture. By transforming and combining these factors, we can manipulate grate paintings, such as relighting them and/or obtaining new views, and render new paintings, e.g., ones with Cézanne’s shading and Renoir’s brush strokes.

Automatic Integration of Foliage Into 3D City Models

Timo Ropinski, Frank Steinicke, Jennis Meyer-Spradow and Klaus Hinrichs

In this paper we introduce visualization techniques for massive multimodal texture datasets. These techniques work on registered texture datasets and have been developed with the goal to improve rendering of these datasets when used in virtual landscape or city environments. We briefly discuss how to render these multimodal datasets at interactive frame rates, and we present an interactive treetop extraction technique, which allows to segment treetops and visualize them as 3D objects to increase realism. Furthermore we apply focus+context visualization techniques to support interactive exploration of multimodal texture datasets.

GPU-based Cell Projection for Large Structured Data Sets

André Maximo, Ricardo Marroquim, Ricardo Farias and Claudio Esperanca

We present a practical implementation of a cell projection algorithm for interactive visualization of large volumetric data sets using programmable graphics cards. By taking advantage of the data regularity, we can avoid computing some steps of the original algorithm with no quality loss. Furthermore, performance is increased since more than half the processing time is dedicated only for rendering. We also provide two tools for better user interactivity, one for transfer function editing and another for volume clipping. Our algorithm generates high quality images at a rendering speed of over 5.0 M Tet/s on current graphics hardware.

Real-Time Adaptive Point Splatting for Noisy Point Clouds

Rosen Diankov and Ruzena Bajcsy

Regular point splatting methods have a lot of problems when used on noisy data from stereo algorithms. Just a few unfiltered outliers, depth discontinuities, and holes can destroy the whole rendered image. We present a new multi-pass splatting method on GPU hardware called Adaptive Point Splatting (APS) to render noisy point clouds. By taking advantage of image processing algorithms on the GPU, APS dynamically fills holes and reduces depth discontinuities without loss of image sharpness. Since APS does not require any preprocessing on the CPU and does all its work on the GPU, it works in real-time with linear complexity in the number of points in the scene. We show experimental results on Teleimmersion stereo data produced by approximately forty cameras.

UNCOUPLED PARALLEL VIEW DEPENDANT LEVEL OF DETAIL RENDERING OF BINTREE TRIANGULATIONS

Bernd Biedermann and María Cecilia Rivara

In this paper we present an uncoupled parallel technique for view dependant continuous level of detail rendering of regular height field terrain. To this end, the terrain is globally modelled by a simple bintree patch-based representation of right-triangles, which is adaptively divided into uncoupled meshes for parallel processing. This is easily performed by uncoupling the mesh along the observer line. Each uncoupled mesh is then recursively approximated to the corresponding level of detail in the terrain. Cracks are avoided by constraining the refinement levels at the boundaries of adjacent meshes. The level of detail is created on-the-fly with a low amount of CPU overhead, allowing a good representation of the terrain and high frames per second performance. This implementation shows significant improvements in CPU load and frames per second performance over the serial method when executed on machines with multiple processors.

Real Time Falling Leafs

Pere-Pau Vázquez and Marcos Balsa

There is a growing interest in simulating natural phenomena in computer graphics applications. Animating natural scenes in real time is one of the most challenging problems due to the inherent complexity of their structure, formed by millions of geometric entities, and the interactions that happen within. An example of natural scenario that is needed for games or simulation programs are forests. Forests are difficult to render because the huge amount of geometric entities and the large amount of detail to be represented. Moreover, the interactions between the objects (grass, leafs) and external forces such as wind are complex to model. In this paper we concentrate in the rendering of falling leafs at low cost. We present a technique that exploits graphics hardware in order to render thousands of leafs with different falling paths at real time and low memory requirements.

A Scalable GPU-Based Approach to Shading and Shadowing for Photorealistic Real-Time Augmented Reality

Claus B. Madsen and Rune Laursen

Visually realistic Augmented Reality (AR) entails addressing several difficult problems. The most difficult problem is that of rendering the virtual objects with illumination which is consistent with the illumination of the real scene. The paper describes a complete AR rendering system centered around the use of High Dynamic Range environment maps for representing the real scene illumination. The main contribution lies in a novel, physically-based approach to rendering shadows cast by virtual objects without changing the shadows already present in the images of the real scene. The proposed approach effectively involves real-time estimation of the diffuse albedos of the real scene, and essentially relighting these areas to take virtual shadows into account. Another contribution lies in the fact that the proposed approach is designed to run on graphics hardware and is scalable in the sense that it offers a simple way to balance performance with visual quality.

INTERACTIVE RENDERING OF MULTIPLE SCATTERING IN PARTICIPATING MEDIA USING SEPARABLE PHASE FUNCTION

Zheng Gong, Zhimin Ren, Zhangye Wang and Qunsheng Peng

This paper presents an interactive method to compute multiple scattering in non-homogeneous participating media with an effective phase function approximation. The volume is represented by grids, which allows us to render dynamic scenes. We achieve interactive computation rate by factorizing the phase function in transport equation with singular value decomposition (SVD) and keeping only the first few low-order approximation terms. These terms are paired 2D incident-direction texture maps and 2D outgoing-direction texture maps. The complicated integral calculation of in-scattering in each rendering pass is efficiently approximated by simply retrieving data from textures. Graphics hardware is also employed for on-the-fly computation. Using the proposed algorithm, we demonstrate rendering of multiple scattering in dynamic scenes at interactive rates.

PRESSURE BASED INK DIFFUSION MODEL FOR REAL-TIME SIMULATION OF CHINESE CALLIGRAPHY

ye wang and Jon Rokne

This paper describes a novel approach to simulating Chinese calligraphy for digital image purposes. The approach includes an ink diffusion model, a multi-layer paper model, a brush model, and the simulation of special effects. Special attention is given to the ink diffusion effect which is of importance in Chinese calligraphy. When the ink is deposited onto absorbent paper, it spreads outside the original border of a stroke since the flow of water will transport carbon particles along the capillary tubes found in the paper. The ink flow model is based on a new algorithm simulating dynamic ink diffusion into absorbent paper. In this capillary network based paper model, the pressure at each node can be obtained from Darcy’s law applied to the ink used in the calligraphy on each edge and it is proportional to the density of capillaries. The deposition layer of the paper is furthermore used to simulate the deposition of carbon particles into the paper and it is also used to simulate the washing out effects. Ink effects such as irregular edges and back run effect can also be simulated. The system is efficient and can create realistic Chinese calligraphy in real-time.

Painting lighting and viewing effects

Cindy Grimm and Michael Kowalski

We present a system for painting how the appearance of an object changes under different lighting and viewing conditions. The user paints what the object should look like under different lighting conditions (dark, partially dark, fully lit, {\em etc.}), or different viewing angles, or both. The system renders the object under new lighting conditions and a new viewing angle by combining these paintings. For surfaces without a pre-defined texture map the system can construct texture maps directly from the user's paintings.

FAST DELAUNAY-TRIANGULATED IMPORTANCE-SAMPLED POINT SETS

In this paper, we propose a novel method to generate Delaunay-triangulated point sets from a given density function in 2D. In order to accomplish this, we employ a Penrose-tiling-based importance-sampling strategy, which not only provides a good sampling point pattern with a local blue-noise distribution, but also provides a balanced base geometric structure from which we can efficiently derive a Delaunay triangulation of the underlying point set. We observe linear execution time with respect to the number of points. There are many areas in computer graphics that can benefit from our fast triangulated point set generator. Typical applications include terrain rendering, 3D geometry processing, and image compression.

SIMULATING REACTIVE/PASSIVE POSTURES BY MEANS OF A HUMAN ACTIVE TORQUE HYBRID MINIMIZATION

Inmaculada Rodríguez and Ronan Boulic

In this paper we propose a hybrid approach minimizing the active torque produced by muscles groups at the joint level. The proposed approach is hybrid in the sense that it combines the local knowledge of the external torque induced by external forces such as gravity and exerted force, and the full knowledge of the passive-resistive torque characteristics due to ligaments and connective tissues. The algorithm is exploited within a context of posture adjustment when a muscle group reaches a critical fatigue level. It proposes a target joint state that can be characterized as active or passive. The active solution, if it exists, can be further characterized by a desired degree of active torque amplitude reduction (between 0 and 100%). In any cases at least one passive solution exists; it relies on the passive/resistive torque appearing in the neighbourhood of the joint limits.

Visual Speech Synthesis From 3D Video

James Edge and Adrian Hilton

Data-driven approaches to 2D facial animation from video have achieved highly realistic results. In this paper we introduce a process for visual speech synthesis from 3D video capture to reproduce the dynamics of 3D face shape and appearance. Animation from real speech is performed by path optimisation over a graph representation of phonetically segmented captured 3D video. A novel similarity metric using a hierarchical wavelet decomposition is presented to identify transitions between 3D video frames without visual artifacts in facial shape, appearance or dynamics. Face synthesis is performed by playing back segments of the captured 3D video to accurately reproduce facial dynamics. The framework allows visual speech synthesis from captured 3D video with minimal user intervention. Results are presented for synthesis from a database of 12minutes (18000 frames) of 3D video which demonstrate highly realistic facial animation.

PLAUSIBLE MOTION SIMULATION: INCHWORM VS. ROLLER

Juha Holopainen and Mauno Rönkkö

In this paper, we study the use of a specific particle system in motion simulation. The novelty of the particle system is that, with it, everything is formalized in a systematic and uniform manner using just seven matrices. We illustrate use of the particle system with two entities, Inchworm and Roller. We first describe the entities and their motion on a simple environment, Easy Road. We then examine how Inchworm and Roller adapt to changes in their environment, as we put them onto a more difficult environment, Rough Road. As the main contribution, we demonstrate here that, despite its simplicity, the particle system supports plausible motion simulation, for organic and mechanistic motion. In addition, we show that, as the entities are modeled in a systematic and uniform manner, they are reusable.

Multi-Mode Representation of Motion Data

Björn Krüger, Jochen Tautges and Andreas Weber

We investigate the use of multi-linear models to represent human motion data. We show that naturally occurring modes in several classes of motion can be used to efficiently represent the motions for various animation task, such as dimensionality reduction or synthesis of new motions by morphing. We show that especially for the approximations of motions by few components the reduction based on a multi-linear model might be considerably better than one obtained by principal component analysis (PCA).

Handling Spatial Vagueness in Virtual Agent Control

Spyros Vosinakis, Nikos Pelekis, Yannis Theodoridis and Themis Panayiotopoulos

Virtual agents are an important element of virtual environments that enhance their believability and autonomy. Nowadays, there are a number of architectures and languages for designing virtual agents and scripting their control process, but the majority of them cannot cope with the vagueness that characterizes a designer’s description of a location or direction using natural language, and, thus, fail to demonstrate complex spatial behavior in dynamic environments. In this paper, we propose a framework for handling spatial vagueness in virtual agent control, which uses binary fuzzy relations to represent vague location descriptions and a fuzzy rule-based system for the agent control. We propose a language for programming the agent’s low-level behavior that allows vague locations based on topological and directional spatial relations, and present a case study, in which the proposed framework is successfully used for a virtual agent’s locomotion in a dynamic environment.

EXPLOITING COUPLED JOINTS

Daniel Raunhardt and Ronan Boulic

In this paper we propose a simple model for the coupling behavior of the human spine that is capable of exhibiting anatomically correct motions of the vertebrae in virtual mannequins. Such a model transparently integrates in our inverse kinematics framework as it couples standard swing and revolute joint models. The adjustment of the joints due to the coupling is made with several simple (in)equality constraints, resulting in a reduction of the solution space dimensionality for the inverse kinematics solver. A key benefit is to prevent the inverse kinematics algorithm from providing infeasible postures. We exploit how to apply these simple constraints to the human spine by a strict decoupling of the swing and twist motion of the vertebrae and demonstrate the validity of our approach on various experiments.

THE ART TO KEEP IN TOUCH the "good use" of Lagrange multipliers

Antoine Jonquet, Olivier Nocent and Yannick Remion

Physically-based modeling for computer animation has become a wide spread framework because it allows to produce more realistic motions in less time without requiring the expertise of skilled animators. But, a computer animation is not only a numerical simulation based on classical mechanics since it follows a precise story-line. One common way to define aims in an animation is to add geometric constraints. There are several methods to manage these constraints within a physically-based framework. In this paper, we present an algorithm for constraints handling based on Lagrange multipliers. After few remarks on the equations of motion that we use, we present a first algorithm proposed by Platt. We show with a simple example that this method is not reliable. Our contribution consists in improving this algorithm to provide an efficient and robust method to handle simultaneous and antagonist constraints.

Interactive Deformation and Visualization of large Volume Datasets

Florian Schulze, Katja Buehler and Markus Hadwiger

This paper presents an integrated approach for interactive direct volume deformation and simultaneous visualization. The fundamental requirement is that interactive performance without pre-processing must be achieved for large volume data, where at any time up to one million elements participate in a deformation that is applied interactively by picking and dragging in the 3D view. Current physically-based approaches are still one or two orders of magnitude away from this goal. In contrast, our approach extends the non-physical ChainMail algorithm and combines it with on-the-fly resampling and GPU ray-casting. Special transfer functions assign material properties depending on volume density. The affected subvolume is deformed and resampled onto a rectilinear grid on the CPU, and updates the volume on the GPU where it is rendered using ray-casting. While the deformation is already being displayed, its quality is simultaneously refined via an iterative relaxation procedure executed in a parallel thread.

FACIAL ANIMATION WITH MOTION CAPTURE BASED ON SURFACE BLENDING

Lijia Zhu and Won-Sook Lee

This paper proposes a methodology to reconstruct 3D facial expressions with motion capture data. Feature-point based facial animation provides easy control of expression usually by moving surface points using mathematical deformation. However it does not support the high quality of surface animation of the face where the feature points are not present. In this paper, we focus on animating a 3D facial model using only feature points, but keeping the high quality animation by using an expression databank obtained from surface scanning. Firstly, a facial expression databank is prepared by processing the raw laser-scanned human face data with a consistent parameterization technique. Secondly, sparse motion capture data is obtained using an optical tracking system. Thirdly, guided by the captured MPEG-4 feature point motions, we find the corresponding surface information in the existing examples in the databank by linear combination of them. The optimized blending weights are obtained implicitly by Genetic Algorithms (GA). Finally, the surface blending result is retargeted into the performer’s neutral facial mesh. Consequently, motions of the facial surface points are reconstructed.

PREDICTIVE-SPECTRAL COMPRESSION OF DYNAMIC 3D MESHES

Rachida Amjoun and Wolfgang Straßer

This paper proposes a new compression algorithm for dynamic 3d meshes. In such a sequence of meshes, neighboring vertices have a strong tendency to behave similarly and the degree of dependencies between their locations in two successive frames is very large which can be efficiently exploited using a combination of \emph{Predictive }and \emph{DCT} coders (PDCT). Our strategy gathers mesh vertices of similar motions into clusters, establish a local coordinate frame (LCF) for each cluster and encodes frame by frame and each cluster separately. The vertices of each cluster have small variation over a time relative to the LCF. Therefore, the location of each new vertex is well predicted from its location in the previous frame relative to the LCF of its cluster. The difference between the original and the predicted local coordinates are then transformed into frequency domain using DCT. The resulting DCT coefficients are quantized and compressed with entropy coding. The original sequence of meshes can be reconstructed from only a few non-zero DCT coefficients without significant loss in visual quality. Experimental results show that our strategy outperforms or comes close to other coders.

Acoustical Diffraction Modeling for Interactive Virtual Environments

Bill Kapralos, Michael Jenkin and Evangelos Milios

Since the dimensions of many of the objects/surfaces encountered in our daily lives are within an order of magnitude as the wavelength of audible sounds, diffraction is an elementary means of sound propagation. Despite its importance in the real-world, diffraction is often overlooked by acoustical modeling methods intended for virtual environment/reality applications leading to a decrease in immersion or presence. This paper describes an acoustical diffraction method that is based on the Huygens-Fresnel principle. The methods is simple and efficient allowing it to be incorporated in interactive virtual environments. Experimental results are presented that illustrate the effectiveness of the method and its conformance to theoretical diffraction models.

A Framework for Interactive GPU-Supported Rendering and Styling of Virtual Hair

Burkhard Wuensche and Rui Zhang

The interactive styling and rendering of virtual hair is essential for creating realistic looking human avatars for use in computer games, virtual worlds, and movie special effects. Hair models can contain tens of thousands of hair strands and hence it is important to develop techniques to modify the hair in a realistic fashion and to render it at interactive frame rates. In this paper we present a GPU-based framework for styling and rendering of virtual hair. We use wisps to represent basic units of hair strands and present an improved statistical model for hair wisp generation which allows the creation of smooth and fringy styles. Fast modelling is achieved by using create, edit, delete and copy and paste operations for key strands representing wisp. The styling process is simplified by using a local coordinate system for hair strands in order to define preferred styling (brushing) directions. Fast photo-realistic rendering is achieved by using the latest GPU functionalities for both the light reflection calculation and the shadow generation of hair strands. We also propose a new method for real-time antialiasing using GPU programming.

Using Augmented Reality for Real-Time Visualization of Tactile Health Examination

Gennadiy Nikishkov and Tomokazu Tsuchimoto

An augmented-reality approach to real-time visualization of tactile data with projection on a human organ is presented. A target procedure is breast cancer examination with a tactile sensor. The special tactile sensor is used to measure tissue stiffness values at discrete points of the examined organ. We developed an augmented reality system which integrates the tactile sensor, a head-mounted display with a small video camera, and a notebook computer. Tracking of the human organ and the tactile sensor is based on optical markers. One marker is attached to the patient's body and is used to track the spatial position of the human organ. Another marker is placed on the tactile sensor. Registering space positions of this marker with respect to the first marker allows determination of the shape of the human organ for subsequent data visualization. Results of stiffness measurements are depicted as semi-transparent three-dimensional objects projected on the patient's body. Different visualization techniques are employed depending on the amount of data and user preference. Experiments with sensing tactile data and its visualization for a variety of objects, including a silicon breast model and animals organs, have been performed. It was found that the developed augmented-reality system is useful in medical-physics measurements.

PRACTICAL DESIGN AND IMPLEMENTATION OF A CAVE SYSTEM

Achille Peternier, Sylvain Cardin, Frédéric Vexo and daniel Thalmann

CAVE systems are nowadays one of the best Virtual Reality (VR) immersive devices available for render-ing virtual environments. Unfortunately, such kind of hardware is extremely expensive, complex and cum-bersome, thus limited in its spread. Several cheaper solutions already exist, but they implement usually only a subset of features of a professional CAVE (like a reduced number of wall-displays, no stereographic at all or with a low-quality support, low brightness, etc.). In this paper we describe how we have built a low cost CAVE with four screens (three walls and a floor), stereographic rendering and user tracking by only using hardware commonly available on the market and free software, we show the different solutions and work-around we implemented to solve the problems we encountered and we conclude with a fair evaluation of our system by using two applications we developed with it.

Visualizing Collapsible 3D Data in a Hybrid GIS

Stephen Brooks and Jacqueline Whalley

In this paper we present a comprehensive set of advancements to our unique hybrid Geographical Information System (GIS). Although many existing commercial 3D GIS systems offer 2D views they are typically isolated from the 3D view in that they are presented in a separate window. Our system is a novel hybrid 2D/3D approach that seamlessly integrates 2D and 3D views of the same data. In our interface, multiple layers of information are continuously transformed between the 2D and 3D modes under the control of the user, directly over a base-terrain. In this way, our prototype GIS allows the user to view the 2D data in direct relation to the 3D view within the same window. In this work we progress the concept of a hybrid GIS by presenting a set of expanded capabilities within our distinctive system. These additional facilities include: landmark layers, 3D point layers, and chart layers, the grouping of multiple hybrid layers, layer painting, the merging of layer controls and consistent zooming functionality.

Improved meshless deformation techniques for real-time interactive collaborative environments

Alex Henriques and Burkhard Wuensche

Meshless deformation based on shape matching is a new technique for simulating deformable objects without requiring mesh connectivity information. The approach focuses on speed, ease of use and stability at the expense of physical accuracy. In this paper we introduce improvements to the technique that increase physical realism and make it more suitable for use in interactive real-time environments such as games and virtual surgery applications. We also present intuitive real-time interaction techniques for picking, pushing and cutting objects simulated using meshless deformation based on shape matching. For deformable collision detection and response, we present a new method for surface meshes based on previous volumetric methods.

3D Spatial Data Mining on Document Sets for the Discovery of Failure Causes in Complex Technical Devices

Timo Götzelmann, Knut Hartmann, Andreas Nürnberger and Thomas Strothotte

The retrospective fault analysis of complex technical devices based on documents emerging in the advanced steps of the product life cycle can reveal error sources and problems, which have not been discovered by simulations or other test methods in the early stages of the product life cycle. This paper presents a novel approach to support the failure analysis through (i) a semi-automatic analysis of databases containing product-related documents in natural language (\eg problem and error descriptions, repair and maintenance protocols, service bills) using information retrieval and text mining techniques and (ii) an interactive exploration of the data mining results. Our system supports visual data mining by mapping the results of analyzing failure-related documents onto corresponding 3D~models. Thus, visualization of statistics about failure sources can reveal problem sources resulting from problematic spatial configurations.

SOME ASSEMBLY REQUIRED: Effectiveness Of Interactive 3D Graphics on Mobile Devices for Object Assembly

A study was conducted to explore the effectiveness of interactive 3D graphics on a mobile device to present instructions for an assembly task: building four Lego models of varying complexity. Our results show significant improvement in assembly correctness and time to complete the assembly when subjects used an interactive presentation compared to a non-interactive one. The study also explored the intuitive notion of 3D object complexity and compared perceived object complexity with experimentally measured values.

FEATURE-POINT DRIVEN 3D EXPRESSION EDITING

I-Chen Lin, Chii-Yuan Chuang, Yung-Sheng Lo and George Lin

Producing a life-like 3D facial expression is usually a labor-intensive process. In movie and game industries, motion capture and 3D scanning techniques, acquiring motion data from real persons, are used to speed up the production. However, acquiring dynamic and subtle details, such as wrinkles, on a face are still difficult or expensive. In this paper, we propose a feature-point-driven approach to synthesize novel expressions with details. Our work can be divided into two main parts: acquisition of 3D facial details and expression synthesis. 3D facial details are estimated from sample images by a shape-from-shading technique. While employing relation between specific feature points and facial surfaces in prototype images, our system provides an intuitive editing tool to synthesize 3D geometry and corresponding 2D textures or 3D detailed normals of novel expressions. Besides expression editing, the proposed method can also be extended to enhance existing motion capture data with facial details.

Interactive Collision Detection for Free-Viewpoint Video

Bert De Decker, Tom Mertens and Philippe Bekaert

We present a novel way of interacting with a virtual 3D scene in the context of free-viewpoint video. Using a multi-camera setup, our technique detects collisions between virtual objects and real objects, including people. We perform collision computations directly on the image data, as opposed to reconstructing the full geometry of the subject. This reduces implementation complexity, and moreover, yields interactive performance. We demonstrate the effectiveness of our technique by incorporating it in a rigid body simulation. The subject can interact with virtual objects and observe his or her actions while being able to adjust the viewpoint, all in real-time.

MOBILE CHASE - Towards a Framework for Location-Based Gaming

Mirko Fetter, Markus Etz and Heiko Blechschmied

Pervasive Gaming and Location-based Games in particular have gained more and more attention recently. Researchers from a variety of fields, media artists, mobile service providers as well as the entertainment industry all seem to have their specific interests in this area. Today already a couple of different games exist from basic applications that are already available to the consumer market to bleeding edge research projects. In this paper we introduce a framework for Location-based Gaming that on the one hand helps with the development of market ready games. On the other hand it serves as a toolkit for researchers aiming to rapidly develop Location-based Games, not having to deal with implementation details far away from their research interests in order to focus on their specific research aspects.

STABLE HAPTIC RESPONSE FOR COMPLEX INTERACTIONS

JOSUNE HERNANTES, DIEGO BORRO and ALEJANDRO GARCÍA-ALONSO

Haptic technology is quite recent and therefore in many cases it is difficult to simulate real contacts or interactions with a high sensation of realism. Collision response methods that calculate the force-feedback tend to cause haptic instabilities when the normal direction changes abruptly. In consequence, collision or contact events are often difficult to render properly in sharp corners by means of haptic devices. This paper describes a collision response method which not only provides users with a stable force feedback, but also a comfortable and convincing haptic interaction. The experimental results show that this approach leads to a smoother force evolution which manages to avoid discontinuities and enhances the quality in the interaction with corners.

A Virtual Reality System for Medical Imaging

Giuseppe De Pietro, Ivana Marra and Carmela Vanzanella

In recent years significant advances in the field of Immersive Medical Imaging Analysis have made it essential to many disciplines related to medicine, such as radiology, neurology, cardiology, radiotherapy. Starting from three dimensional (3D) image datasets produced by computed tomography (CT) or magnetic resonance imaging (MRI) scans, for these areas it can be very useful to use simulated visualizations of human organs. By viewing the inside of anatomical structures and interacting with them, doctors can better understand the data of interest for medical training, surgical simulations, examination and diagnosis aims.In this paper we present a virtual reality system, that involves new software components based on top of open-source and cross-platform libraries; it consists of a set of services implementing immersive 3D navigation and interaction of virtual representations of human organ structures, starting from DICOM data. These features have been developed to be integrated into a medical imaging component-based architecture, which at present is under development.

REAL-TIME DEFORMABLE OBJECTS FOR COLLABORATIVE VIRTUAL ENVIRONMENTS

Selcuk Sumengen, Mustafa Tolga Eren, Selim Balcisoy and Serhat Yesilyurt

This paper presents a method for physical simulation of deformable closed surfaces over a network, which is suitable for realistic interactions between users and objects in a collaborative virtual environment (CVE). CVE's are extensively used for training, design and gaming for several years. To demonstrate a deformable object in a CVE, we employ a real-time physical simulation of a uniform-tension-membrane, based on linear finite-element-discretization of the surface yielding a sparse linear system of equations, which is solved using the Runge-Kutta Fehlberg method. The proposed method introduces an architecture that distributes the computational load of physical simulation between each participant. Our approach requires a uniform-mesh representation of the simulated structure; therefore we designed and implemented a re-meshing algorithm that converts irregularly triangulated genus zero surfaces into a uniform triangular mesh with regular connectivity. The strength of our approach comes from the subdivision methodology that enables to use multi-resolution surfaces for graphical representation, physical simulation, and network transmission, without compromising simulation accuracy and visual quality.

A METHOD TO SYNTHESIZE THREE-DIMENSIONAL FACIAL MODEL BASED ON THE INFORMATION OF WORDS EXPRESSING FACIAL FEATURES

Futoshi Sugimoto and Masahide Yoneyama

Our aim is to synthesize faces based on freely-elicited expressions by expanding the range of words describing the shape of facial features to include abstract or metaphorical expressions. We realize this by defining the synthesizing process of a human face as a mapping from a word space to a physical model space. The use of whole words existing in the word space has made it possible to synthesize a human face based on freely-elicited expressions.

OpenCrimeScene Review Log - Interaction Log in a Virtual Crime Scene Investigation Learning Environment

Angela Brennecke, Stefan Schlechtweg and Thomas Strothotte

In this paper we present a concept for visualising an interaction log in a virtual crime scene investigation learning environment called OpenCrimeScene. The interaction log shall be used for reviewing the user’s behaviour after having examined the virtual crime scene. Furthermore, it shall serve as a valuable discussion background when it comes to self or teacher’s control. As initial point the log visualisation therefore has to present an instant overview of user activity combined with significant user interactions. In order to keep track of the causal relations and the sequence of events we here focus on visualising time by combining different significant character positions in one image. This paper primarily aims at giving an overview of our system and introduces the review log as well as its first visualisation form.

DEVELOPMENT OF RUTOPIA 2 VR ARTWORK USING NEW YGDRASIL FEATURES

Daria Tsoupikova and Alex Hill

Ygdrasil is a programming framework used by artists and computer scientists worldwide to create networked multi-user virtual reality (VR) worlds and tele-immersive art projects. Features have been added recently to improve the rendering capabilities and extend the user's creative control. This paper describes the development of the VR art project Rutopia 2 using these new features.

Towards intellgent VR: Multi-Layered Semantic Reflection for Intelligent Virtual Environments

Marc Erich Latoschik and Christian Fröhlich

This paper introduces semantic reflection, a novel design concept for intelligent Virtual Environments (IVEs). SCIVE—a simulation core particularly aimed at IVE development—provides semantic reflection as a major principle on multiple layers: It’s architecture grants semantic driven uniform access to low-level simulation core logic, to specific simulation modules’ application definitions, as well as to high-level semantic environ- ment descriptions. It additionally provides a frame to conveniently interconnect various simulation modules, e.g., for graphics, physics, audio, haptics, or AI etc. SCIVE’s Knowledge Representation Layer’s base for- malism provides the central organizing structure for the diverse modules’ data representations. It allows bidi- rectional knowledge driven access between the modules since their specific data structures and functions are transitively reflected by the semantic layer. Hence SCIVE preserves, integrates and provides unified access to the development paradigms of the interconnected modules, e.g., scene graph metaphors or field route concepts etc. well known from todays Virtual Reality systems. SCIVE’s semantic reflection implementation details are illustrated following a complex example application. We illustrate how semantic reflection and modularity support extensibility and maintainability of VR applications, potential for automatic systemconfigurationandoptimization, as well as the base for comprehensive knowledge driven access for IVEs.

Porting Legacy Applications to Immersive Virtual Environments: A Case Study

Kenny Gruchalla, Jonathan Marbach and Mark Dubin

Immersive virtual environments are becoming increasingly common, driving the need to develop new software or adapt existing software to these environments. We discuss some of the issues and limitations of porting an existing molecular graphics system, PyMOL, into an immersive virtual environment. Presenting macromolecules inside an interactive immersive virtual environment may provide unique insights into molecular structure and improve the rational design of drugs that target a specific molecule. PyMOL was successfully extended to render molecular structures immersively; however, elements of the legacy interactive design did not scale well into three-dimensions. Achieving an interactive frame rate for large macromolecules was also an issue. The immersive system was developed and evaluated on both a shared-memory parallel machine and a commodity cluster.

COMPARATIVE EXPERIMENT OF BODY-POSITION BASED NAVIGATION IN IMMERSIVE VIRTUAL ENVIRONMENTS

Kikuo Asai

Navigation is one of the basic human-computer interactions in virtual environments, and the technique should be easy to use, cognitively simple, and uncumbersome. However, the interaction in immersive virtual environments requires a factor of the sense of immersion as well as efficiency. We have proposed a body-position based navigation technique that drives a viewpoint with extension and bending of the arms, rotating both arms, and standing on tiptoes and bending the knees. Using various parts of the body may help to enhance the sense of immersion in the virtual environment. Depth images obtained from a polynocular stereo machine are used for tracking the 3D positions of the arms and head of the user in an immersive projection display. We evaluated the body-position system in experiments in which participants performed fly-through tasks in a 3D space, and compared the effectiveness of the body-position system with that of a joystick and a hand-arm gesture interface. The results of the experiment showed that the body-position system was advantageous on moving around at large areas instead of efficiency or accuracy of viewpoint control in virtual environments.

A Fuzzy Path Finding Algorithm Based on Artificial Potential Fields

Razvan Tudor Tanasie and Dorian Cojocaru

This paper will present a path finding algorithm based on a fuzzy system for a time dependent problem. The fuzzy system uses singleton fuzzifier, product inference engine and center average defuzzifier. The algorithm considers a static environment (the obstacles are not moving). The inputs are the space map, which is split in squares (considered as the basic motion elements), the initial position of the wanderer and the positions of the obstacles and the target. The algorithm computes, if possible, a path from the position of the wanderer to the target, and uses the artificial potential field approach to compute weights for each of the possible future positions of the wanderer. It is implemented in Microsoft Visual C++ and uses DirectX 9.0 libraries.

3D TECHNIQUES TO CREATE INTERACTIVE VIRTUAL MUSEUMS: THE STATE OF THE ART IN THE EPOCH NOE

Pitzalis Denis, Rodriguez Echavarria Karina, Lahanier Christian, Arnold David B., Pillay Ruven and Aitken Genevieve

Information and multimedia, such as images or 3D models, stored in databases are very important to preserve the information about historical artefacts and works of art. Nevertheless, the potential of digital content in databases is not fully exploited until it is used to create interactive ways to communicate to non CH specialist the significance that these objects have. 3D virtual environments are a suitable mechanism for giving context to, otherwise isolated, pieces of information. To achieve this, different techniques for 3D acquisition, integration and visualisation must work together in order to create 3D interactive virtual environments which are engaging and accessible for the visitors of a museum. In this paper we will describe the state of the art of the techniques for achieving this type of environments within the partners of the EPOCH Network of Excellence.

CONTROLLING A VIRTUAL BODY BY THOUGHT IN A HIGHLY-IMMERSIVE VIRTUAL ENVIRONMENT - A Case Study in using a Brain-Computer Interface in a Virtual-Reality Cave-like

Doron Friedman

A brain-computer interface (BCI) can arguably be considered the ultimate user interface, where humans operate a computer using thought alone. We have integrated the Anon-BCI into a highly immersive Cave-like system. In this paper we report a case study where three participants were able to control their avatar using only their thought. We have analyzed the participants’ subjective experience using an in-depth qualitative methodology. We also discuss some limitations of BCI in controlling a virtual environment, and interaction design decisions that needed to be made.

MECHANICAL PERFORMANCE OF A MANIPULATOR IN VIRTUAL REALITY SYSTEMS

Jose San Martin and Gracian Trivino

Frequently, the human interface of a virtual reality system includes a 3D manipulator. In order to optimize the use of this device, the designer must take into account its mechanical characteristics. An obvious design criterion consists of maximizing the coincidence between the application 3D space and the physical volume where the manipulator provides its maximum performance. This paper explains in detail the analysis of manipulability for the PHANToM OMNi haptic device including the study of the manipulability distribution into its real workspace boundaries. As result of this study we will define a measure of the quality of the device placement inside the virtual reality system platform. We apply this measure for designing the mechanical configuration of a simulator for Minimally Invasive Arthroscopic Surgery.