Distribution ray tracing uses Monte Carlo integration to solve the rendering equation. This technique was introduced by Cook et. al, and was notable because of its simplicity and its ability to simulate areal luminaires, camera lens e ects, motion blur, and imperfect specular re ection[5]. Distribution

The handling of highly complex 3D scenes is one of the major challenges in computer graphics. Several data structures were proposed in the past to address this problem. Many of these schemes are only suited for specific spatial distribution of objects in 3D space, making it difficult for a developer to select the appropriate data structure for the scene and/or application. Further, the selection of initialization parameters is typically a non-trivial task. Using a ray casting environment this paper presents an algorithm that automatically builds a hybrid data structure combining bounding volume hierarchies and uniform spatial subdivisions for a given scene. Our data structure is built by first creating a cost function based volume hierarchy, subsequently detecting regions of uniformly distributed objects using the scene hierarchy, and, as the last step, locally integrating uniform spatial subdivisions into the scene tree. We will show that the data structure can be built with low costs, both with regard to space and to run-time. Finally, we present rendering times that demonstrate the usefulness of the new approach compared to standard techniques by comparing run-time efficiency. The memory requirements of the new data structure are, on average, linear in the number of scene objects. Further applications of the hybrid approach are also proposed. Key Words: spatial subdivision, geometry structuring, bounding volume hierarchies, photorealistic rendering, graphics architecture. 1.

The ability to perform efficient collision detection is essential in virtual reality environments and their applications, such as walkthroughs. In this paper we re-explore a classical structure used for collision detection -- the binary space partitioning tree. Unlike the common approach, which attributes equal likelihood to each possible query, we assume events that happened in the past are more likely to happen again in the future. This leads us to the definition of self-customized data structures. We report encouraging results obtained while experimenting with this concept in the context of self-customized bsp trees. Keywords: Collision detection, binary space partitioning, self-customization. 1 Introduction Virtual reality refers to the use of computer graphics to simulate physical worlds or to generate synthetic ones, where a user is to feel immersed in the environment to the extent that the user feels as if "objects" seen are really there. For example, "objects" should m...

In this paper an acceleration method for finding the nearest ray--object intersection for ray tracing purposes is presented. We use the concept of BSP trees enriched by rope trees. These are used to accelerate the traversal of the BSP tree. We give a comparison of experimental results between the technique based on BSP tree and uniform spatial subdivision. Key words: spatial subdivision, BSP tree, spatial data structures, rope trees, ray tracing. 1 Introduction Ray tracing is a well--known rendering technique for producing realistic images that simulates well specular surfaces. The main drawback of this algorithm is its rather big computational complexity, that disallows its interactive use. The problem has been focused a great deal of research interest in the past leading to some practical techniques to accelerate the basic algorithm. The principal expense of ray tracing is the determination of the closest ray--object intersection for a given ray and a set of objects. This problem i...

We describe two new techniques of ray shooting acceleration that exploit the traversal coherence of a spatial hierarchy. The first technique determines a sequence of adjacent leaf--cells of the hierarchy that is pierced by all rays contained within a certain convex shaft. This sequence is used to accelerate ray shooting for all remaining rays within the shaft. The second technique establishes a cut of the hierarchy that contains nodes where the hierarchy traversal can no longer be predetermined for all rays contained within a given shaft. This cut is used to initiate the traversal for all remaining rays contained in the shaft. The description of the methods is followed by results evaluated by their practical implementation. Keywords: ray shooting, ray casting, BSP tree, traversal coherence, hidden surface removal. 1.

Recently developed interactive ray tracing systems combine the high-performance of todays CPUs with new algorithms and implementations to achieve a flexible and highperformance rendering system offering high-quality, interactive 3D graphics. However, due to its history in off-line rendering, interactive ray tracing has been limited to static scenes and simple walkthroughs. However, in order to become truly interactive ray tracing must support dynamic scenes efficiently.

Density estimation on the tangent plane (DETP) is a density estimation technique for global illumination. This technique is based on Photon Maps and provides increased accuracy when the surfaces are not locally smooth or continuous. However, the performance of the technique is limited by the large number of ray-disc intersections needed. Some optimizations which increase the performance of DETP have been devised. The first optimization works by creating a set of candidate rays for each radiance calculation. The second optimization uses spatial indexing of the discs around the radiance calculation points. An analytical study of the order of complexity of the algorithms, as well as an heuristic study of the calculation time for the different values of the parameters involved, has been performed. Some rules are given in order to identify the most suitable optimization for a given radiance calculation.

The short overview of several acceleration techniques for ray tracing algorithm is presented. The aim of this paper is to show how combination of both software improvements and hardware support can help to achieve higher speed of rendering. Such simple but memory and time consuming algorithm like ray tracing is an ideal case for understanding of typical speeding up methods used in three dimensional computer graphics. Most of presented approaches thus have more general validity. Keywords computer graphics, rendering, ray tracing, speeding up, parallel algorithm. 1 Introduction The ray tracing algorithm is one of the most popular methods in computer graphics and probably everybody at least partially experienced in 3D graphics knows it. The simplicity of a ray tracing (its basic version respectively) has attracted many people to implement it with the highest possible efficiency. But still no real--time ray tracer is known to author, although research papers on software improvements a...

One of the most fundamental concepts in computer graphics is binary space subdivision. In its purest form, this concept leads to binary space partitioning trees (BSP trees) with arbitrarily oriented space partitioning planes. In practice, however, most algorithms use kd-trees—a special case of BSP trees that restrict themselves to axis-aligned planes—since BSP trees are believed to be numerically unstable, costly to traverse, and intractable to build well. In this paper, we show that this is not true. Furthermore, after optimizing our general BSP traversal to also have a fast kd-tree style traversal path for axis-aligned splitting planes, we show it is indeed possible to build a general BSP based ray tracer that is highly competitive with state of the art BVH and kd-tree based systems. We demonstrate our ray tracer on a variety of scenes, and show that it is always competitive with—and often superior to—state of the art BVH and kd-tree based ray tracers.

The ray-scene intersection test is the most costly process when a scene is rendered. This process may be improved using any strategy to be able to speed-up it. Generally, any strategy used is based on the building of a spatial indexing in the scene domain or in the rays domain. In this paper, an acceleration techniques formalization is proposed. This formalization allows an optimizer to be specified according to the spatial index used. Furthermore, a formalization of optimizer composition is presented. Finally, we present an expression which allows to compare several optimizers, and select the one with best performances. This formalization is based on the graphics objects theory and claims to be a generalization to all optimizers which use spatial indexing. Keywords: Graphics object theory, spatial indexing, ray casting, acceleration techniques. 1