In this paper we introduce a new approach to controlling error in hierarchical clustering algorithms for radiosity. The new method ensures that just enough work is done to meet the user's quality criteria. To this end the importance of traditionally ignored visibility error is identified, and the concept of features is introduced as a way to evaluate the quality of an image. A methodology to evaluate error based on features is presented, which leads to the development of a multi-resolution visibility algorithm. An algorithm to construct a suitable hierarchy for clustering and multi-resolution visibility is also proposed. Results of the implementation show that the multiresolution approach has the potential of providing significant computational savings depending on the choice of feature size the user is interested in. They also illustrate the relevance of the featurebased error analysis. The proposed algorithms are well suited to the development of interactive lighting simulation syste...

Abstract: 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. 1

This paper describes algorithms that allow multiple hierarchical radiosity solvers to work on the same radiosity solution in parallel. We have developed a system based on a group iterative approach that repeatedly: 1) partitions patches into groups, 2) distributes a copy of each group to a slave processor which updates radiosities for all patches in that group, and 3) merges the updates back into a master solution. The primary advantage of this approach is that separate instantiations of a hierarchical radiosity solver can gather radiosity to patches in separate groups in parallel with very little contention or communication overhead. This feature, along with automatic partitioning and dynamic load balancing algorithms, enables our implemented system to achieve significant speedups running on moderate numbers of workstations connected by a local area network. This system has been used to compute the radiosity solution for a very large model representing a five floor building with furni...

Contents Acknowledgements 3 Foreword 9 1 Introduction 11 2 PreviousWork 14 2.1 The Radiosity equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1.1 Rendering Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1.2 Rendering Equation for diffuse surfaces . . . . . . . . . . . . . . . . . . . . 16 2.1.3 The Radiosity system of equations . . . . . . . . . . . . . . . . . . . . . . . 17 2.1.4 Two forms of the Form Factor integral . . . . . . . . . . . . . . . . . . . . . 18 2.1.5 The Form Factor integral as a contour integral . . . . . . . . . . . . . . . . 18 2.1.6 Differential area to area Form Factor . . . . . . . . . . . . . . . . . . . . . . 19 2.2 Computing the Form Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.2.1 Deterministic numerical solutions . . . . . . . . . . . . . . . . . . . . . . . . 22 2.3 Monte Carlo evaluation of the Form Factor integral . . . . . . . . . . . . . . . . .

The ability to perform interactive walkthroughs of global illumination solutions including glossy effects is a challenging open problem. In this paper we overcome certain limitations of previous approaches. We first introduce a novel, memory- and compute-efficient representation of incoming illumination, in the context of a hierarchical radiance clustering algorithm. We then represent outgoing radiance with an adaptive hierarchical basis, in a manner suitable for interactive display. Using appropriate refinement and display strategies, we achieve walkthroughs of glossy solutions at interactive rates for non-trivial scenes. In addition, our implementation has been developed to be portable and easily adaptable as an extension to existing, diffuse-only, hierarchical radiosity systems. We present results of the implementation of glossy global illumination in two independent global illumination systems.

An algorithm for simulating diffuse interreflection in complex three dimensional scenes is described. It combines techniques from hierarchical radiosity and multiresolution modelling. A new face clustering technique for automatically partitioning polygonal models is used. The face clusters produced group adjacent triangles with similar normal vectors. They are used during radiosity solution to represent the light reflected by a complex object at multiple levels of detail. Also, the radiosity method is reformulated in terms of vector irradiance and power. Together, face clustering and the vector formulation of radiosity permit large savings. Excessively fine levels of detail are not accessed by the algorithm during the bulk of the solution phase, greatly reducing its memory requirements relative to previous methods. Consequently, the costliest steps in the simulation can be made sub-linear in scene complexity. Using this algorithm, radiosity simulations on scenes of one mi...

Figure 1: Left: An office which receives light through a window on the right wall while the rest of the scene is lit indirectly. Our method avoids explicit visibility computation and our GPU implementation enables the manipulation of the light or objects at 9 frames per second (fps). This includes four iterations of radiance and antiradiance per frame. Center: We can also include an animated character; indirect light is updated at 7.8 fps. Right: Our method naturally handles glossy surfaces; the glossy bust is lit indirectly, since the light points at the ceiling. We reformulate the rendering equation to alleviate the need for explicit visibility computation, thus enabling interactive global illumination on graphics hardware. This is achieved by treating visibility implicitly and propagating an additional quantity, called antiradiance, to compensate for light transmitted extraneously. Our new algorithm shifts visibility computation to simple local iterations by maintaining additional directional antiradiance information with samples in the scene. It is easy to parallelize on a GPU. By correctly treating discretization and filtering, we can compute indirect illumination in scenes with dynamic objects much faster than traditional methods. Our results show interactive update of indirect illumination with moving characters and lights.

Traditionally, Radiosity algorithms have been restricted to scenes made from planar patches. Most algorithms for computing form factors and the subdivision criterion for hierarchical methods implicitly assume planar patches. In this paper, we present a new radiosity algorithm that is solely based on simple geometric information about surface elements, namely their bounding boxes and cone of normals. Using this information allows to compute efficient error bounds that can be used for the subdivision oracle and for computing the energy transfer. Due to the simple interface to geometric objects, our algorithms not only allows for computing illumination on general curved surfaces, but it can also directly be applied to a hierarchy of clusters. Several examples demonstrate the advantages of the new approach.

When a rendering algorithm has created a pixel array of radiance values the task of producing an image is not yet completed. In fact, to visualize the result the radiance values still have to be mapped to luminances, which can be reproduced by the used display. This step is performed with the help of tone reproduction operators. These tools have mainly been applied to still images, but of course they are just as necessary for walkthrough applications, in which several images are created per second. In this paper we illuminate the physiological aspects of tone reproduction for interactive applications. It is shown how tone reproduction can also be introduced into interactive radiosity viewers, where the tone reproduction continuously adjusts to the current view of the user. The overall performance is decreased only moderately, still allowing walkthroughs of large scenes. 1. Introduction Physically based rendering can be split into three phases. First a model of the virtual worl...

This paper presents a new method for the production of view-independent global illumination solutions of complex static environments. A key innovation of this new approach is its decomposition of the problem into a loosely coupled sequence of simple modules. This approach decouples the global energy transport computation from the construction of the displayable shaded representation of the environment. This decoupling eliminates many constraints of previous global illuminationapproaches, yieldingaccurate solutions for environments with non-diffuse surfaces and high geometric complexity. Our algorithm produces a view-independent display mesh that represents the irradiances on surfaces in a form that allows direct display of the shaded surface. Most traditional radiosity algorithms also use a computational mesh to represent intermediate results in the light transport calculation (e.g., the piecewise-constant global solution of Smits et al. [17]). Typically, a single representation is used for both the computational and display meshes (e.g. the static mesh used by Neumann et al. [11] and the adaptive mesh used by Teller et al. [18]). Very few display mesh solutions have been produced for environments with more than a few thousand initial surfaces. The only implementation we are aware of that has produced a display mesh for more than 10,000 initial surfaces is the system by Teller et al. [18], which was run on a model with approximately 40,000 initial surfaces. Teller et al. argue that the reason for these surprisingly small limits is the high memory overhead of the data structures associated with the computational mesh. To solve this problem, we draw on an observation by Lischinski et al. [10], that the computational mesh and the display mesh have different purposes and ch...