Skip to main content

Über dieses Buch

The book contains the proceedings of the 8th Eurographics Rendering Workshop, which took place from 16th to 18th June, 1997, in Saint Etienne, France. After a series of seven successful events the workshop is now well established as the major international forum in the field of rendering and illumination techniques. It brought together the experts of this field. Their recent research results are compiled in this proceedings together with many color images that demonstrate new ideas and techniques. This year we received a total of 63 submissions of which 28 were selected for the workshop after a period of careful reviewing and evaluation by the 27 mem­ bers of the international program committee. The quality of the submissions was again very high and, unfortunately, many interesting papers had to be rejected. In addition to regular papers the program also contains two invited lectures by Shenchang Eric Chen (Live Picture) and Per Christensen (Mental Images). The papers in this proceedings contain new research results in the areas of Finite-Element and Monte-Carlo illumination algorithms, image-based render­ ing, outdoor and natural illumination, error metrics, perception, texture and color handling, data acquisition for rendering, and efficient use of hardware. While some contributions report results from more efficient or elegant algo­ rithms, others pursue new and experimental approaches to find better solutions to the open problems in rendering.



Polyhedral Geometry and the Two-Plane Parameterization

Recently the light-field and lumigraph systems have been proposed as general methods of representing the visual information present in a scene. These methods represent this information as a 4D function of light over the domain of directed lines. These systems use the intersection points of the lines on two planes to parameterize the lines in space. This paper explores the structure of the two-plane parameterization in detail. In particular we analyze the association between the geometry of the scene and subsets of the 4D data. The answers to these questions are essential to understanding the relationship between a lumigraph, and the geometry that it attempts to represent. This knowledge is potentially important for a variety of applications such as extracting shape from lumigraph data, and lumigraph compression.
Xianfeng Gu, Steven J. Gortler, Michael F. Cohen

Image-based Rendering with Controllable Illumination

A new image-based rendering method, based on the light field and Lumigraph system, allows illumination to be changed interactively. It does not try to recover or use any geometrical information (e.g., depth or surface normals) to calculate the illumination, but the resulting images are physically correct. The scene is first sampled from different viewpoints and under different illuminations. Treating each pixel on the back plane of the light slab as a surface element,the sampled images are used to find an apparent BRDF of each surface element. The tabular BRDF data of each pixel is further transformed to the spherical harmonic domain for efficient storage. Whenever the user changes the illumination setting, a certain number of views are reconstructed. The correct user perspective view is then displayed using the texture mapping technique of the Lumigraph system. Hence, the intensity, the type and the number of the light sources can be manipulated interactively.
Tien-Tsin Wong, Pheng-Ann Heng, Siu-Hang Or, Wai-Yin Ng

View-based Rendering: Visualizing Real Objects from Scanned Range and Color Data

Modeling arbitrary real objects is difficult and rendering textured models typically does not result in realistic images. We describe a new method for displaying scanned real objects, called view-based rendering. The method takes as input a collection of colored range images covering the object and creates a collection of partial object models. These partial models are rendered separately using traditional graphics hardware and blended together using various weights and soft z-buffering. We demonstrate interactive viewing of real, non-trivial objects that would be difficult to model using traditional methods.
Kari Pulli, Hugues Hoppe, Michael Cohen, Linda Shapiro, Tom Duchamp, Werner Stuetzle

Applying Shape from Lighting Variation to Bump Map Capture

We describe a system for capturing bump maps from a series of images of an object from the same view point, but with varying, known, illumination. Using the illumination information we can reconstruct the surface normals for a variety of, but not all, surface finishes and geometries. The system allows an existing object to be rerendered with new lighting and surface finish without explicitly reconstructing the object geometry.
Holly Rushmeier, Gabriel Taubin, André Guéziec

Interactive Common Illumination for Computer Augmented Reality

The advent of computer augmented reality (CAR), in which computer generated objects mix with real video images, has resulted in many interesting new application domains. Providing common illumination between the real and synthetic objects can be very beneficial, since the additional visual cues (shadows, interreflections etc.) are critical to seamless real-synthetic world integration. Building on recent advances in computer graphics and computer vision, we present a new framework to resolving this problem. We address three specific aspects of the common illumination problem for CAR: (a) simplification of camera calibration and modeling of the real scene; (b) efficient update of illumination for moving CG objects and (c) efficient rendering of the merged world. A first working system is presented for a limited sub-problem: a static real scene and camera with moving CG objects. Novel advances in computer vision are used for camera calibration and user-friendly modeling of the real scene, a recent interactive radiosity update algorithm is adapted to provide fast illumination update and finally textured polygons are used for display. This approach allows interactive update rates on mid-range graphics workstations. Our new framework will hopefully lead to CAR systems with interactive common illumination without restrictions on the movement of real or synthetic objects, lights and cameras.
George Drettakis, Luc Robert, Sylvain Bougnoux

An Error Metric for Monte Carlo Ray Tracing

A method is presented for characterizing the error in Monte Carlo realistic image synthesis calculations. An error metric has been developed that can be used to control the variance in the final picture by choosing both the number of rays to be cast into the image plane and the number of rays to be spawned at each bounce in the environment. The method provides specific guidance in how to apply Russian Roulette and splitting at each level of the ray tree. An initial implementation of the method has been done to test the theory and to illustrate its mechanics.
Mark R. Bolin, Gary W. Meyer

Selective Culling of Discontinuity Lines

In recent years discontinuity meshing has become an important part of mesh-based solutions to the global illumination problem. Application of this technique accurately locates all radiance function discontinuities in the scene and is essential for limiting visual artifacts. In an environment containing m edges there are O(m 2) D 1 and D 2 discontinuities. With a typical scene this can result in many thousands of discontinuity lines being processed. We review existing methods for reducing these lines and introduce an improved, perception based metric for determining which discontinuities are important and which can be safely ignored. Our results show that a 50% or more reduction in the number of discontinuity lines can be achieved with a corresponding reduction in general mesh complexity, with little or no perceptible change to the rendered result.
David Hedley, Adam Worrall, Derek Paddon

Reconstructing the Visual Field of Compound Eyes

Realistic image synthesis research involves the simulation of visible light propagation within an environment with a view to computing an image that evokes a similar visual response to that perceived by an observer of that environment. This has involved the use of both sophisticated camera response and human visual system response models. In this paper we investigate the application of invertebrate vision models, in particular the vision from compound eyes. We use Apis meliferra (or honey bee) as a case study, constructing a geometric model of the retinal structure of the eye and applying psychophysical data obtained from physiological, morphological and behavioural studies regarding spectral sensitivity and spatial acuity in order to reconstruct an image approximating that perceived by a bee. The algorithm is general and can be adapted to other invertebrate families.
Steven Collins

Interactive Rendering of Globally Illuminated Glossy Scenes

Global illumination simulates all transfers of light in a scene. The results of the simulation are then used to generate photo-realistic images. Scenes with diffuse surfaces only can be displayed in real-time using the results of radiosity methods. Images of scenes with more general surfaces are created with methods based on ray tracing but do not achieve interactive frame rates.
This paper presents a new algorithm for the display of globally illuminated scenes at interactive speeds. A photon tracing phase computes an approximation to the global illumination. The rendering phase splats the contribution of each photon hit onto the corresponding surface taking reflectance properties and viewing direction into account. Results demonstrate that this method allows to render images of globally illuminated scenes with glossy surfaces at interactive frame rates.
Wolfgang Stürzlinger, Rui Bastos

Incremental Updates to Scenes Illuminated by Area Light Sources

An object space algorithm for computing shadows in dynamic scenes illuminated by area light sources is presented. A mesh with the shadow boundaries as well as other discontinuities in the illumination function, is built in a pre-processing stage and updated on-line after any interaction resulting in a change in the scene geometry. The mesh on each polygon is a 2D BSP tree stored in a winged edge data structure. To accelerate the mesh construction a number of new ideas are employed: sorting of the polygons in respect to the area source, the shadow overlap cube, BSP tree merging of the shadows. In addition a method for dynamically changing the BSP representation of the mesh and quickly identifying the vertices requiring intensity computations was developed. Preliminary experimental results indicate the strength and the potential of this method.
Y. Chrysanthou, M. Slater

Progressive Previewing of Ray-Traced Images Using Image-Plane Discontinuity Meshing

This paper presents a new method for progressively previewing a ray-traced image while it is being computed. Our method constructs and incrementally updates a constrained Delaunay triangulation of the image plane. The points in the triangulation correspond to all of the image samples that have been computed by the ray tracer, and the constraint edges correspond to various important discontinuity edges in the image. The triangulation is displayed using hardware Gouraud shading, yielding a piecewise-linear approximation to the final image. Texture mapped surfaces, as well as other regions in the image that are not well approximated by linear interpolation, are handled with the aid of hardware texture mapping.
Frédéric P. Pighin, Dani Lischinski, David Salesin

An adaptive representation of spectral data for reflectance computations

This paper deals with the representation of spectral data so as to control the colorimetric error committed during rendering computations. These data are projected on a set of hierarchical basis functions called scaling functions leading to a representation by means of binary trees. An adaptive algorithm is proposed in which refinement and merge steps managed by an estimation of the error made in the X Y Z color space allows to control the representation of spectra
Gilles Rougeron, Bernard Péroche

A Geometry Dependent Texture Generation Framework for Simulating Surface Imperfections

To model surface imperfections and weathering, we propose a two-step texture generation framework in between manual texture synthesis and automatic physical simulation. Although the pattern of blemishes looks random, the systematic and geometry dependent nature of the underlying distribution is still observable. A distribution of tendency (potential to contain blemishes) is modeled in the first step, which includes user control and geometric information. The second generates and distributes an irregular blemish pattern according to the modeled tendency distribution. As examples we model three common surface imperfections; dust accumulation, patina and peeling.
Tien-Tsin Wong, Wai-Yin Ng, Pheng-Ann Heng

Nailboards: A Rendering Primitive for Image Caching in Dynamic Scenes

This paper proposes a simple augmentation to texture mapping hardware which produces the correct depth buffer content and hence correct visibility when replacing complex objects by partially transparent textured polygons. Rendering such polygons exploits frame-to-frame coherence in image sequences of dynamic scenes.
Correct depth values are obtained by keeping a small depth delta for every texel which represents the texel’s deviation from the textured polygon. The polygon’s depth values are modified at every pixel to match the depicted object’s geometry.
Gernot Schaufler

Filtered Local Shading in the Wavelet Domain

Many global illumination algorithms generate directionally-and positionally-varying radiance data that then need to be somehow re-sampled and used for final shading. This operation should filter all light over the incident hemisphere through the BRDF to generate an accurate image. This can be done analytically for simple BRDFs, such as Lambertian or Phong-like BRDFs, but becomes more difficult in the presence of a general BRDF.
This paper presents an efficient method to calculate the reflected light in a given direction by filtering over all incident light directions. The method exploits wavelet representations of incident light and of the BRDF to compute the total relfected light in a given direction. For efficiency the incident light is restricted to a Haar transformed representation, while the BRDF can be represented and compressed with any appropriate basis. The method can be used with any system that can generate projections of incident light fields onto Haar wavelet bases.
Paul Lalonde, Alain Fournier

An Empirical Comparison of Progressive and Wavelet Radiosity

This paper presents a comparison of basic progressive and wavelet radiosity algorithms. Several variants of each algorithm were run on a set of scenes at several parameter settings, and results were examined in terms of their error, speed, and memory consumption. We did not compare more advanced variations such as clustering or discontinuity meshing. Our results show that progressive radiosity with substructuring works fairly well for all scenes. Among wavelet methods, the Haar basis works best, while higher order methods suffer because of extreme memory consumption and because poor visibility handling causes discontinuous, blocky shadows.
Andrew J. Willmott, Paul S. Heckbert

Hierarchical Radiosity On Curved Surfaces

Incorporating curved objects into a hierarchical radiosity system typically bears a great disadvantage: the initial tessellation already needs a large number of small polygons because further mesh enhancement is impossible. We will show that improvements in rendering speed and quality can be made by extending the planar meshing of the refinement step to an object-specific subdivision scheme. While keeping the number of input polygons extremely low an arbitrary accuracy of the solution can be obtained.
Stephan Schäfer

A Hierarchical Subdivision Algorithm for Stochastic Radiosity Methods

The algorithm proposed in this paper uses a stochastic approach to incrementally calculate the illumination function over a surface. By tracking the illumination function at different levels of meshing resolution, it is possible to get a measure for the quality of the current representation, and to adaptively subdivide in places with inadequate accuracy. With this technique a hierarchical mesh that is based on the stochastic evaluation of global illumination is generated.
Robert F. Tobler, Alexander Wilkie, Martin Feda, Werner Purgathofer

Bidirectional Radiosity

In this paper we present a new algorithm for solving the global illumination problem, based on the mathematical framework resulting from the dual set of equations that describe light transport in a three-dimensional environment. The proposed method is a finite element algorithm and propagates radiance as well as potential, thereby focusing on the rapid and efficient computation of the flux emitted by selected patches. We will show that the method takes into account all possible light paths, and that a faster solution can be obtained compared to other radiosity algorithms.
Ph. Dutré, Ph. Bekaert, F. Suykens, Y. D. Willems

Hierarchical Visibility in Terrains

This paper describes a hierarchical visibility technique that significantly accelerates terrain rendering. With this technique, large parts of the terrain that are hidden from the viewpoint are culled, thus avoiding the expense of uselessly sending them down the graphics pipeline (only to find in the z—buffer step that they are hidden). The hierarchical visibility technique has been implemented in a multiresolution terrain rendering algorithm and experimental results show very large speedups in some situations.
A. James Stewart

Hierarchical Lighting Simulation for Outdoor Scenes

Lighting algorithms for outdoor scenes suffer from the sheer geometric and lighting complexity of such environments. In this paper we introduce an effi­cient, hierarchical solution to the problem of outdoor illumination. Data structures and sampling algorithms are presented, permitting the integration of complex and natural objects in a hierarchical radiosity simulation system. This new approach allows the hierarchical simulation of radiant energy exchanges in outdoor scenes for the first time, including terrain and botanical models as well as sunlight and skylight. This is accomplished by providing the necessary tools to treat terrain meshes as a hierarchy of light-exchanging objects, as well as an efficient hierar­chical representation for the sky dome. In addition, refinement criteria are adapted to the particular characteristics of natural lighting. Results of our implementation are presented including naturally-lit images of terrain-maps, trees and buildings.
Katja Daubert, Hartmut Schirmacher, François X. Sillion, George Drettakis

Plane-Parallel Radiance Transport for Global Illumination in Vegetation

This paper applies plane-parallel radiance transport techniques to scattering from vegetation. The leaves, stems, and branches are represented as a volume density of scattering surfaces, depending only on height and the vertical component of the surface normal. Ordinary differential equations are written for the multiply scattered radiance as a function of the height above the ground, with the sky radiance and ground reflectance as boundary conditions. They are solved using a two-pass integration scheme to unify the two-point boundary conditions, and Fourier series for the dependence on the azimuthal angle. The resulting radiance distribution is used to precompute diffuse and specular “ambient” shading tables, as a function of height and surface normal, to be used in rendering, together with a z-buffer shadow algorithm for direct solar illumination.
Nelson Max, Brett Keating, Curtis Mobley, En-Hua Wu

Lighting Reconstruction Using Fast and Adaptive Density Estimation Techniques

Monte Carlo (MC) photon shooting approach is becoming an important global illumination technique in research and commercial applications. In this work, we focus on the problem of lighting reconstruction for planar surfaces. Our contribution is in the development of new, efficient photon density estimation techniques. We formulate local error measures of lighting reconstruction which under some reasonable constraints (discussed below) imposed on the lighting function that behave like the actual error. The minimization of our error estimates is very fast for planar surfaces and usually leads to a better quality lighting result than traditional methods. Also, the local error estimation offers more information than global error measures usually provided by MC solvers, which are not good predictors of image quality. We compare the actual error resulting from various techniques, and evaluate the visual appearance of the reconstructed lighting.
Karol Myszkowski

A neuro-evolutionary unbiased global illumination algorithm

In this paper we present a two pass unbiased global illumina­tion rendering algorithm. First pass calculations are done shooting rays from light sources and storing directional information in a growing adap­tive neural gas structure. The second pass is a ray tracing process which uses this information to create an evolving population of rays that tend to optimally sample their surroundings. Finally, weighted Monte Carlo integration is used with a dynamic Voronoi diagram to reduce uncer­tainty in the solution.
Eduardo Bustillo

Improved Irradiance Computation by Importance Sampling

In this paper we consider the task of refining a low-resolution radiosity solution by using final gather. This final gather can be done by using Monte-Carlo methods. We show how to built an adequate probability density function (pdf) such that the involved variance is greatly reduced and thus we can reduce the number of samples to take, keeping the error. The proposed pdf uses information gathered during the computation of low-resolution radiosity. This data includes an approximation to the distribution of irradiance landing on a patch and coming from other patches. Once this distribution is (approximately) known, we can use it to built an importance-based pdf for final gather, such that just a few tens of samples can be taken to accurately compute irradiance.
C. Ureña, J. C. Torres

Adaptive Sampling and Bias Estimation in Path Tracing

One of the major problems in Monte Carlo based methods for global illumination is noise. This paper investigates adaptive sampling as a method to alleviate the problem. We introduce a new refinement criterion, which takes human perception and limitations of display devices into account by incorporating the tone-operator. Our results indicate that this can lead to a significant reduction in the overall RMS-error, and even more important that noisy spots are eliminated. This leads to a very homogeneous image quality. As most adaptive sampling techniques our method is biased. In order to investigate the importance of this problem, a nonparametric bootstrap method is presented to estimate the actual bias. This provides a technique for bias correction and it shows that the bias is most significant in areas with indirect illumination.
Rasmus Tamstorf, Henrik Wann Jensen

Application of Rendering Techniques to Monte-Carlo Physical Simulation of Gas Diffusion

Gas diffusion in a microporous media is strongly analogous to light transport in a scene. Interaction of light with surfaces has a counterpart in Knudsen diffusion and light dispersion in participating media is very similar to gaseous bulk diffusion. On behalf of this fact, a Monte-Carlo simulation method has been designed to compute gas diffusivities and related properties in 3D images of porous media, making use of techniques originated in the domain of image rendering, such as tesselation, spatial partitioning, distance images, optimized intersection tests as well as in the domain of physical simulations, such as microscopic/macroscopic random walk algorithms. Putting all these methods together results in a very efficient algorithm allowing to determine accurately transport properties in real porous media from imaging techniques such as X-ray tomography.
Philippe Blasi, Bertrand Le Saëc, Gérard Vignoles

Global Illumination Techniques for the Simulation of Participating Media

This paper surveys global illumination algorithms for environments including participating media and accounting for multiple scattering. The objective of this survey is the characterization of those methods: Identification of their base techniques, their assumptions, limitations and range of utilization. To this end, the algorithms are grouped into functional categories and each method is briefly reviewed, with a discussion of its complexity and its pros and cons. We finish by discussing some applications as well as remaining areas for investigation.
Frederic Pérez, Xavier Pueyo, François X. Sillion

Global Illumination for Professional 3D Animation, Visualization, and Special Effects

The simulation of global illumination has expanded from pure research to industrial software development. This has created new challenges to meet the requirements of professional 3D rendering software users in digital media authoring and visualization. This paper describes these requirements and discusses the approach the company mental images has taken towards fulfilling them.
Per H. Christensen


Weitere Informationen