5TH EUROGRAPHICS WORKSHOP ON VIRTUAL ENVIRONMENTS

Session A1: Levels of Detail

Validity-Preserving Simplification of Very Complex Polyhedral Solids, C. Andújar, D. Ayala, P. Brunet (Spain)
Abstract: In this paper we introduce the Discretized Polyhedra Simplification (DPS), a framework for polyhedra simplification using space decomposition models. The DPS is based on a new error measurement and provides a sound scheme for error-bounded, geometry and topology simplification while preserving the validity of the model. A method following this framework, Direct DPS, is presented and discussed. Direct DPS uses an octree for topology simplification and error control, and generates valid solid representations. Our method is also able to generate approximations which do not interpenetrate the original model, either being completely contained in the input solid or bounding it. Unlike most of the current methods, restricted to triangle meshes, our algorithm can deal and also produces faces with arbitrary complexity.

View-Dependent Topology Simplification, J. El-Sana, A. Varshney (NY, USA)
Abstract: We propose a technique for performing view-dependent simplifications for level-of-detail-based renderings of complex models. Our method is based on exploiting frame-to-frame coherence and is tolerant of various commonly found degeneracies in real-life polygonal models. The algorithm proceeds by preprocessing the input dataset into a binary tree of vertex collapses. This tree is used at run time to generate the triangles for display. Dependencies to avoid mesh foldovers in manifold regions of the input object are stored in the tree in an implicit fashion. This obviates the need for any extra storage for dependency pointers and suggests a potential for applica-tion to external memory prefetching algorithms. We also propose a distance metric that can be used to unify the geometry and genus simplifications with the view-dependent parameters such as viewpoint, view-frustum, and local illumination.

Adaptive tessellation of connected primitives for interactive walkthroughs in complex industrial virtual environments, M. Krus, P. Bourdot, A. Osorio, F. Guisnel, G. Thibault (France)
Abstract: Geometrical primitives used in virtual environments are converted to an important amount triangles at rendering time. The meshes of the resulting simplifications usualy introduce discontinuities between neighbouring object. We extend a simple adaptive tessellation method which adapts the amount of triangles to the viewing conditions with connection information to ensure that the meshes of connected primitives remain continuous. An ergonomical study has validated this approach for applications using virtual environments.

Industrial Presentation

A comparison of projection technologies
Andreas v. Erdmannsdorff, BARCO (Germany)
Abstract: The presentation discusses the criteria for the selection of the best suited projection technology for a given application. It shows that each technology has its strengths and weaknesses, so the decision which technology to use is strongly dependent of the application. Therefor also the criteria are discussed with respect to the applications. At the end a tool to help to find the best projection technology for the application is developed.

Session A2: Tracking

An Optical Tracking System for VR/AR-Applications, K. Dorfmueller (Germany)
Abstract: In this paper, an optical tracking system is introduced for the use within Virtual and Augmented Reality applications. The system uses retroflective markers which are attached to a special designed interaction device. The construction of the device allows us to gather six degrees of freedom. In order to achieve high tracking precision we introduce a calibration algorithm which results in sub-pixel accuracy and is therefore well applicable within Augmented Reality scenarios. Further the algorithm for calculating the pose of a rigid body is described. Finally, the optical tracking system is evaluated in regard to its accuracy.

The integration of optical and magnetic tracking for multi-user augmented reality, T. Auer, S. Brantner, A. Pinz (Austria)
Abstract: Multi-user augmented reality requires excellent registration for all users, which cannot be achieved by magnetic trackers alone. This paper presents a new approach combining magnetic and optical tracking. The magnetic tracker is used to coarsely predict the positions of land- marks in the camera image. This restricts the search area to a size which can be managed close to real-time. This new hybrid tracking system out-performs a calibrated magnetic tracker in terms of position, orientation, and jitter.

An Optically Based Direct Manipulation Interface for Human-Computer Interaction in an Augmented World, G. Klinker, D. Stricker, D. Reiners (Germany)
Abstract: Augmented reality (AR) constitutes a very powerful three-dimensional user interface for many "hands-on" application scenarios in which users cannot sit at a conventional desktop computer. To fully exploit the AR paradigm, the computer must not only augment the real world, it also has to accept feedback from it. Such feedback is typically collected via gesture languages, 3D pointers, or speech input - all tools which expect users to communicate with the computer ABOUT the work at a meta-level rather than just letting them pursue their task. When the computer is capable of deducing progress directly from changes in the real world, the need for special abstract communication interfaces can be reduced or even eliminated. In this paper, we present an optical approach for analyzing and tracking users and the objects they work with. In contrast to emerging workbench and metaDESK approaches, our system can be set up in any room after quickly placing a few known optical targets in the scene. We present three demonstration scenarios to illustrate the overall concept and potential of our approach and then discuss the research issues involved.

Session A3: Rendering of VEs

Improving the Illumination Quality of VRML 97 Walkthrough via Intensive Texture Usage, C. Kardassevitch, J.P. Jessel, M. Paulin, R. Caubet (France)
Abstract: In this paper, we introduce a pipeline, dedicated to global illumination and walkthrough of a VRML 97 scene. This pipeline use intensively and exclusively textures to represent light, and is entirely guided by them. We will show how reversing the classic rendering pipeline, allows to privilege high frequency information (direct illumination), and to accelerate the complete lighting process (global illumination). Finally, we will present the filtering and reconstruction methods used to improve the rendering visual quality.

Fast Walkthroughs with Image Caches and Ray Casting, M. Wimmer, M. Giegl, D. Schmalstieg (Austria)
Abstract: We present an output-sensitive rendering algorithm for accelerating walkthroughs of large, densely occluded virtual environments using a multi-stage Image Based Rendering Pipeline. In the first stage, objects within a certain distance are rendered using the traditional graphics pipeline, whereas the remaining scene is rendered by a pixel based approach using an Image Cache, horizon estimation to avoid calculating sky pixels, and finally, ray casting. The time complexity of this approach does not depend on the total number of primitives in the scene. We have measured speedups of up to one order of magnitude.

Using Virtual Environments to Enhance Visualization, D.R.S. Boyd, J.R. Gallop, K.E.V. Palmen, R.T. Platon, C.D. Seelig (UK)
Abstract: Within the EU ESPRIT demonstrator project VIVRE, a commercial virtual environment system has been used to create a user-centred interaction environment for two commercial data visualization systems. The project has exploited mechanisms provided by all three systems to incorporate new user-developed functionality. This includes a communication library, geometry conversion utilities and user control of visualization parameters from within the virtual environment. The project is assessing the degree to which this extended interactive capability results in greater benefits for commercial users.

Session B1: Distributed Environments

Semantic Behaviours in Collaborative Virtual Environments, Emmanuel Frécon, Gareth Smith (Sweden and UK)
Abstract: Scripting facilities within a collaborative virtual environment (CVE) allows animation and behaviour to be added to otherwise static scenes. This paper describes the integration of the TCL scripting language into an existing CVE, and describes the advantages gained by such a marriage. Further, we describe how high-level semantic behaviours can be readily introduced into cooperative applications. They benefit from the scripting language to provide an abstraction over application development and be exploited to drastically reduce network traffic.

A Distributed Device Diagnostics System Utilizing Augmented Reality and 3D Audio, R. Behringer, S. Chen, K. Wang, V. Sundareswaran (CA, USA)
Abstract: Augmented Reality brings technology developed for virtual environments into the real world. This technology can be used to provide instructions for routine maintenance and error diagnostics of technical devices. We are developing such a system which utilizes Augmented Reality techniques to indicate problem zones of a device directly into the view of the user. The system can overlay 3D rendered objects, animations, and text annotations onto the video image of a known object. This allows the user to localize problems of the device with the actual device in his view. The user can query the status of device components using a speech recognition system. The response is given as a animation of the relevant device module and/or as auditory cues using spatialized 3D audio. The position of the user relative to the device is tracked by a computer vision tracking system. The diagnostics system also allows the user to leave spoken annotations attached to device modules for other users to retrieve. The system is implemented on a distributed network of PCs, utilizing standard commercial off-the-shelf components (COTS).

Texture-based Volume Visualization for Multiple Users on the World Wide Web, K. Engel, T. Ertl (Germany)
Abstract: We present a texture-based volume visualization tool, which permits remote access to radiological data and supports multi-user environments. The application uses JAVA and the Virtual Reality Modeling Language (VRML), thus it is platform-independent and able to use fast 3D graphics acceleration hardware of client machines. The application allows the shared viewing and manipulation of three-dimensional medical volume datasets in a heterogeneous network. Volume datasets are transferred from a server to different client machines and locally visualized using a JAVA-enabled web-browser. In order to reduce network traffic, a data reduction and compression scheme is proposed. The application allows view dependent and orthogonal clipping planes, which can be moved interactively. On the client side, the users are able to join a visualization session and to get the same view onto the volume dataset by synchronizing the viewpoint and any other visu-alization parameter. Interesting parts of the dataset are marked for other users by placing a tag into the visualization. In order to support collaborative work users communicate with a chat applet, which we provide, or by using any existing video conferencing tool.

Session B2: Systems and Applications

PVR An Architecture for Portable VR Applications, R. van Liere, J. D. Mulder (The Netherlands)
Abstract: Virtual reality shows great promise as a research tool in computational science and engineering. However, since VR involves new interface styles, a great deal of implementation effort is required to develop VR applications. In this paper we present PVR; an event-based architecture for portable VR applications. The goal of PVR is to provide a programming environment which facilitates the development of VR applications. PVR differentiates itself from other VR toolkits in two ways: First, it decouples the coordination and management of multiple data streams from actual data processing. This simplifies the programmer's task of managing and synchronizing the data streams. Second, PVR strives for portability by shielding low-level device specific details. Application programmers can take full advantage of the underlying hardware while maintaining a single code base spanning a variety of input and output device configurations.

Rapid Development of VRML Content via Geometric Programming, A. Paoluzzi, S. Francesi, S. Portuesi, M. Vicentino (Italy)
Abstract: This paper aims to show that a functional design language can be used as a general-purpose VRML generator. The PLaSM language, which allows for algebraic computations with geometric shapes and maps, has been recently extended with non geometric attributes like colors, lights and textures. PLaSM is so used in the paper both to develop some geometric tools, including Bezier manifolds of any dimension and degree, the n-th derivative of any parametric curve and "Bezier stripes" of small width, as well as to quickly implement a quite complex mountain landscape. Customized PLaSM applications may generate fully parametrized virtual worlds starting from small data files or streams.

Augmented Reality, the other way around, D. Verna, A. Grumbach (France)
Abstract: Given a situation of Virtual Reality in which an operator interacts with a virtual environment through a manipulation device such as a robot, this paper shows that the notion of Augmented Reality, along with its applications, has been developed in a biased way: mostly in destination to the operator, and at the perception level. This demonstration is achieved through a model describing a situation of Tele-Operation, on which we represent major cases of Augmented Reality encountered in recent applications. By taking advantage of the symmetry of the model, we are able to show how Augmented Reality can be seen "the other way around", that is, in destination to the environment, and at the action level.

Session B3: Interaction

Evaluation of Interaction Techniques on the Virtual Workbench, R. van de Pol, W. Ribarsky, L. Hodges, F. Post (The Netherlands, USA)
Abstract: This paper evaluates interaction methods within the general framework of navigation, selection, and manipulation. It considers large display environments and, in particular, the virtual workbench, comparing this system with HMD and CAVE systems. The paper addresses three issues: (a) identifying the characteristics that set the workbench apart from other virtual environments; (b) determining types, organization, and examples of interaction techniques: (c) evaluating how these techniques perform on the workbench to determine which perform best. The evaluations are based on an extensive set of user observations. Also discussed are some problems that stereoscopic display coupled with interaction bring out.

A General Framework for Cooperative Manipulation in Virtual Environments, D. Margery, B. Arnaldi, N. Plouzeau (France)
Abstract: Whereas cooperation and collaboration have become two popular words in virtual reality, the problem of cooperative manipulation has been mainly left aside due to the great number of other challenges facing anyone trying to setup multi-user worlds. We define cooperative manipulation as a situation where two or more users interact on the same object in a concurrent but cooperative way. The focus of this paper is to describe an experiment whose goal was to experiment problems specific of cooperative manipulation setups. Those problems include synchronizing user's input over the network, mapping user's input into a meaningful 3-D movement thanks to what we call a model of activity and giving him relevant visual information. In this paper, we present a general framework able to take into account these problems. It is compatible with physically simulated objects and has been implemented using Java, VRML and a distributed approach.

Occlusion in Collaborative Augmented Environments, A. Fuhrmann, G. Hesina, F. Faure, M. Gervautz (Austria)
Abstract: Augmented environments superimpose computer enhancements on the real world. Such augmented environments are well suited for collaboration of multiple users. To improve the quality and consistency of the augmentation the occlusion of real objects by computer-generated objects and vice versa has to be implemented. We present methods how this can be done for a tracked user's body and other real objects and how irritating artifacts due to misalignments can be reduced. Our method is based on simulating the occlusion of virtual objects by a representation of the user modeled as kinematic chains of articulated solids. Registration and modeling errors of this model are being reduced by smoothing the border between virtual world and occluding real object. An implementation in our augmented environment and the resulting improvements are presented.