A. Glassner, Space subdivision for fast ray tracing. IEEE Computer Graphics & Applications, Oct. 1984.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....using pointers in the hierarchy. When adding animation to the scene, whole or parts of the acceleration data structures will most likely have to be reconstructed between frames. Depending on the amount of changes in the scene, this could seriously stress the reconstruction phase when using octrees [7], uniform grids [8] recursive grids [9] hierarchical grids [10, 11, 12] BSP trees [13, 14] and bounding volume hierarchies (BVHs) 15, 16] In our benchmark, we have excluded the animation of light sources, because it simplified our animations, and the same stress can be achieved by animated ....

....data structures 3 each frame. This is easily done for translations, rotations, and scalings, but often other kinds of less organized animation is used, and then this approach cannot be used. Since, to our knowledge, all currently available types of acceleration schemes for ray tracing (e.g. [7, 8, 9, 12, 14, 16]) must rebuild their efficiency structures for such animations, this will be a serious stress on all ray tracing algorithms. Stress 3 : Teapot in the stadium The teapot in the stadium problem [18] refers to when a small detailed object (teapot) is located in a relatively large surrounding ....

[Article contains additional citation context not shown here]

Glassner, Andrew S., "Space Subdivision for Fast Ray Tracing", IEEE Computer Graphics and Applications, vol. 4, no. 10, pp. 15--22, October 1984.

.... drag and gravity . As the force generators are created procedurally it is easy to create a great number of them. This can lead to performance problems since calculating the dynamics takes more time as the number of forces increases. To counter this problem we have used octree optimization [Gla84]: The scene is split hierarchically to boxes and the hierarchy is updated progressively (as was done by O Brien [OBr01] Since most of the forces have limited bounding box we can avoid calculating the forces that could not affect particles that are in another box. Whether this optimization does ....

Glassner A., Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications. 4(10), pages 15-22, 1984

....More over bounding spheres share the same disadvantages as bounding boxes. For the reasons above, we adapted a variant of hierarchical oct tree spatial sub division for partitioning object space. Spatial decomposition of object space using adaptive oct trees is also popular for ray tracing [Glassner84] Glassner uses this method to reduce the number of objects tested with each ray for expensive ray object intersection tests. In collision detection, it will be useful in reducing even more expensive geometric intersection tests of each object with other objects in the scene which are usually in ....

....the face without resorting to intersections. A 2 d version of this method has been proposed: the digitization of straight line on 2 d raster grid using Digital Differential Analyzer (DDA) Rogers85] But for 3 d identification of voxels pierced by a straight line, the method described by [Glassner84] can be used with back face culling technique [Glassner90] for efficiency. But real gains could be obtained by extending 2 DDA concept to digitization of straight line in 3 d space for efficient identification of cells in spatial decomposition in ray tracing [Fujimoto86] An elegant algorithm is ....

Andrew S. Glassner, "Space subdivision for fast ray tracing", IEEE CG&A, vol.4, No.10, pp.15-22, October, 1984.

....by excluding various objects from consideration, sight unseen. Lin [17] mentions two widely used techniques: hierarchical bounding, which partitions a scene into a tree of enclosing volumes [11, 14] and space subdivision, which divides the objects among a tree of uniformly sized spatial cells [10]. Only objects within the cells or bounding volumes through which a ray passes are tested against that ray. Scherson and Caspary [24] analyze the performance of such algorithmic improvements in depth; the performance gains realized are significant. Parallel ray tracing is done either ....

AS Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, 1984.

....by excluding various objects from consideration, sight unseen. Lin [10] mentions two widely used techniques: hierarchical bounding, which partitions a scene into a tree of enclosing volumes [6,9] and space subdivision, which divides the objects among a tree of uniformly sized spatial cells [5]. Only objects within the cells or bounding volumes through which a ray passes are tested against that ray. Scherson and Caspary [14] analyze the performance of such algorithmic improvements in depth; the performance gains realized are significant. Parallel ray tracing is done either ....

Andrew Glassner, "Space subdivision for fast ray tracing," IEEE Computer Graphics and Applications, 4(10):15--22, 1984.

....tracing based radiosity [16] 13] a large part of the computation time is spent in the calculation of potential or actual intersections between rays and objects in the scene. In the last decade, numerous optimization techniques have been proposed (bounding volumes [17] 12] slabs [11] octrees [6], grids [5] macro regions [4] five dimensional subdivisions [1] pyramids [15] hybrid decompositions [14] to reduce the number of unneeded intersection computations. This paper presents a new space partitioning scheme for ray tracing based algorithms which divides a given region of the scene ....

....Scene Dependent Space Partitioning: In the second family, regions are computed more or less independently from the position and orientation of the objects. More precisely, the whole scene is subdivided either uniformly by rectangular grids [5] or spherical pyramids [15] or adaptatively by octrees [6], non uniform grids [10] or macro regions [4] Note that a fivedimensional partitioning has also been proposed [1] in which the scene is subdivided according to the three axis defining a position and the two angles defining a direction. Nevertheless this technique is somehow different from the ....

A.Glassner, Space Subdivision for Fast Ray Tracing, Computer Graphics and Applications, v4, n10, p15-22, 1984.

....RAY TRACING USING K D TREES Donald Fussell and K. R. Subramanian The University of Texas at Austin Austin, Texas 78712 1188 TR 88 07 March 1988 Fast Ray Tracing Using K D Trees Donald Fussell K. R. Subramanian Abstract A hierarchical search structure for ray tracing based on k d trees is introduced. This data structure can handle the variety of surfaces commonly used in computer graphics. Algorithms to build and traverse this ....

....object in a scene. This process has to be done several times per pixel, depending on the effects we are trying to model (reflection, transmission, shadows, penumbras etc) Considerable effort has gone into making this process very efficient. Bounding volumes [18] 22] 23] 14] search structures [8] 9] 13] item buffers [22] shadow buffers [11] and ray coherence [1] 121120] all have drastically reduced this cost. Bounding volumes are advantageous because they provide a quick non intersection test for objects. Using tight bounding volumes improves the chances of intersecting the object inside ....

[Article contains additional citation context not shown here]

Andrew S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15-22, October 1984.

....box. The most natural would be to allocate four arrays for each of above values. Nevertheless, it turns out that processor works better with data, which is stored in the nearby memory blocks. Most effective will be to create a single array of the following structs: struct Trglnfo int vertind[3]; int plnind; UINT flags; float box[2] 3] There are also other methods of low level optimizations. General guideline on such type of optimizations can be found in [2] In this article we would like to focus more on the higher level optimizations, in particular on the voxel grid ....

....four arrays for each of above values. Nevertheless, it turns out that processor works better with data, which is stored in the nearby memory blocks. Most effective will be to create a single array of the following structs: struct Trglnfo int vertind[3] int plnind; UINT flags; float box[2] [3]; There are also other methods of low level optimizations. General guideline on such type of optimizations can be found in [2] In this article we would like to focus more on the higher level optimizations, in particular on the voxel grid creation traversing algorithms. 3. VOXELIZATION ....

[Article contains additional citation context not shown here]

A.S.Glassner. "Space subdivision for fast ray-tracing", IEEE C.G. &A., 4(10)pp 15-22 1984.

....acceleration techniques have concentrated on reducing the cost of ray casting by exploiting various kinds of coherence between rays. This has been done by casting bundles of rays, such as beams [41] and cones [3] employing a fixed grid [32] or adaptive 3D spatial hierarchy of scene objects [34] and directional techniques [6] A good summary of these techniques can be found in Arvo and Kirk [7] Besides the above techniques used to accelerate the generation of a single ray traced image, some researchers have addressed the problem of generating ray traced image sequences rapidly for ....

Andrew Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, October 1984.

....the objects that intersect the voxel. When a ray passes through space, only those objects that lie in voxels intersected by the ray must be tested to detect ray object intersections. Depending on the different space subdivision strategies one distinguishes between regular and adaptive techniques [5, 6, 10, 3] (Figure 2) Figure 2. Uniform and adaptive spatial subdivision. Because ray tracing acceleration schemes differ in their relative efficiency from scene to scene, no single technique can be optimal under all circumstances [2] Bounding volume hierarchies are well suited for inhomogeneous scenes, ....

A. S. Glassner. Space Subdivision for Fast Ray Tracing, IEEE Computer Graphics and Applications, 4(10), pp. 15--22, Oct. 1984.

.... characteristics, we now face the question: to subdivide or not to subdivide Traditionally, this question is answered by ad hoc termination criteria (further abbreviated to AHTC) algorithm, which was given with the introduction of ray shooting algorithms based on the ## tree [Kapla85] and octree [Glass84]. The AHTC follows the common sense without thoroughly considering the distribution of objects in the scene: the current node # becomes a leaf when the number of objects intersecting the ##### is lower than or equal to a fixed constant # ### , or its depth #### reaches another fixed constant # ....

Glassner,A Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, Oct. 1984.

....preprocessing and uses linear memory space. Thus instead of implementing the algorithms invented in computational geometry, computer graphics practitioners prefer heuristic ray shooting speed up techniques, including, for example, ffl regular space partitioning [FTK86] AK89] ffl octree [Gla84] AK89] ffl ray coherence methods [OM87] HMSK92] ffl ray classification [Gla84] AK89] ffl Voronoi diagram based space partitioning [M ar95a] These algorithms are usually not analyzed formally. From worst case complexity point of view, these algorithms are even worse than the naive ....

.... algorithms invented in computational geometry, computer graphics practitioners prefer heuristic ray shooting speed up techniques, including, for example, ffl regular space partitioning [FTK86] AK89] ffl octree [Gla84] AK89] ffl ray coherence methods [OM87] HMSK92] ffl ray classification [Gla84] AK89] ffl Voronoi diagram based space partitioning [M ar95a] These algorithms are usually not analyzed formally. From worst case complexity point of view, these algorithms are even worse than the naive solution. However, for the majority of the cases they seem to be better than that. ....

[Article contains additional citation context not shown here]

Andrew S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, 1984.

....data structure. There are two main classes of hierarchies used to order the scene. Object subdivision clusters the objects, recording the space that each object inhabits (hierarchical extent tree, 3] 5] Space subdivision subdivides space, recording the objects that intersect each region ([2] [4] 8] 1] 2.1 Binary Space Partitioning Tree The binary space partitioning (BSP) tree [2] 4] is a common type of space subdivision. Initially, the BSP consists of one node, representing the bounding volume containing all objects of the scene. This volume is divided using a splitting planes ....

....clusters the objects, recording the space that each object inhabits (hierarchical extent tree, 3] 5] Space subdivision subdivides space, recording the objects that intersect each region ( 2] 4] 8] 1] 2. 1 Binary Space Partitioning Tree The binary space partitioning (BSP) tree [2] [4] is a common type of space subdivision. Initially, the BSP consists of one node, representing the bounding volume containing all objects of the scene. This volume is divided using a splitting planes perpendicular to one of the major axes. Each leaf of the BSP contains references to all objects ....

A. S. Glassner. Space subdivision for fast raytracing. IEEE Computer Graphics Applications, 4:15-22, 1984.

No context found.

A. Glassner, Space subdivision for fast ray tracing. IEEE Computer Graphics & Applications, Oct. 1984.

No context found.

A. Glassner, Space subdivision for fast ray tracing. IEEE Computer Graphics & Applications, Oct. 1984.

No context found.

A. S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15-22, Oct. 1984.

No context found.

Andrew S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, October 1984.

No context found.

A. S. Glassner. Space sub-division for fast ray tracing. IEEE Computer Graphics and Applications, October 1984.

No context found.

Andrew S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, October 1984.

No context found.

Glassner,A.: Space Subdivision for Fast Ray Tracing, IEEE CG&A, Vol. 4, No. 10, pp. 15-22, 1984.

No context found.

Andrew S. Glassner. Space Subdivision For Fast Ray Tracing. IEEE Computer Graphics & Applications, 4(10):15--22, October 1984.

No context found.

A.S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, October 1984.

No context found.

A. S. Glassner. Space subdivision for fast ray tracing. IEEE Computer Graphics and Applications, 4(10):15--22, 1984.

No context found.

A. S. Glassner, \Space Subdivision for Fast Ray Tracing," IEEE Computer Graphics and Applications, no. 4, pp. 15-22, Oct. 1984.

No context found.

Glassner,A.S., "Space Subdivision for Fast Ray Tracing," IEEE Computer Graphics and Applications, 4(10), pp. 15-22 (October 1984).

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC