Implementations of Cube-4 on the Teramac custom computing machine

Kanus, U.; Meißner, Michael; Straßer, Wolfgang; Pfister, Hanspeter; Kaufman, Arie; Amerson, R.; Carter, Richard J.; Culbertson, W. Bruce; Kuekes, Philip J.; Snider, Greg

doi:10.1016/s0097-8493(96)00083-0

Cited by 9 publications

(8 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A simulation of the architecture on the HP Teramac configurable custom hardware machine running at about 1 Mhz on a 125 3 dataset with five parallel pipelines achieved a frame rate of 1.6 Hz (Kanus et al, 1997). A VLSI implementation of the architecture scales well even for huge datasets.…”

Section: Real-time Hardwarementioning

confidence: 99%

Real-time volume rendering

Kaufman¹,

Dachille²,

Chen³

et al. 2000

Int. J. Imaging Syst. Technol.

Self Cite

View full text Add to dashboard Cite

With the advent of high‐powered, commodity volume visualization hardware comes a new challenge: effectively harnessing the visualization power to enable greater understanding of data through dynamic interaction. We examine Cube‐4/VolumePro as the latest advance in real‐time volume visualization hardware. We describe tools to utilize this hardware including a software developers' kit, called the Cube Toolkit (CTK). We show how the CTK supports algorithms such as perspective rendering, overlapping volumes, and geometry mixing within volumes. We examine a case study of a virtual colonoscopy application developed using the CTK.© 2000 John Wiley & Sons, Inc. Int J Imaging Syst Technol, 11, 44–52, 2000

show abstract

Section: Real-time Hardwarementioning

confidence: 99%

Real-time volume rendering

Kaufman¹,

Dachille²,

Chen³

et al. 2000

Int. J. Imaging Syst. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Cube-4 is a pipelined scalable volume rendering architecture based on ray casting [4,13,141. Cube-4L is a modification of Cube-4 simplifying it by using an implementation of rayslice-sweeping.…”

Section: The Cube-4l Architecturementioning

confidence: 99%

“…Unfortunately, it arrives there a full beam processing time (b cycles) too late. The Cube-4 architecture handled this by reading the beam as soon as possible, but delaying the processing of the data by b cycles which requires buffering of the whole beam [4,13]. Figure 14 suggests a better approach by shifting the beginning of a beam by one partial beam to the right.…”

Section: Abc Gradient Estimationmentioning

confidence: 99%

A ray-slice-sweep volume rendering engine

Bitter

Kaufman

1997

Proceedings of the ACM SIGGRAPH/EUROGRAPHICS Workshop on Graphics Hardware - HWWS '97

View full text Add to dashboard Cite

Ray-slice-sweeping is a plane sweep algorithm for volume rendering, The compositing buffer sweeps through the volume and combines the accumulated image with the new slice of just-projected voxels. The image combination is guided by sight rays from the view point through every voxel of the new slice. Cube-.#L is a volume rendering architecture which employs a ray-slice-sweeping algorithm. It improves the Cube-4 architecture in three ways. First, during perspective projection all voxels of the dataset contribute to the rendering. Second, it computes gradients at the voxel positions which improves accuracy and allows a more compact implementation, Third, Cube-AL has less control overhead than Cube-C IntroductionVolume visualization is a method of extracting information from volumetric datasets through the use of interactive graphics and imaging, and is concerned with the representation, manipulation and rendering of these datasets [5]. Most commonly, the data represents a continuous 3D function sampled on a ree;ular, rectilinear 3D grid of volume elements called uoxels. Depending on this 3D function (mathematical, MRI, CT, ultrasound, etc.) voxels accordingly represent function value, density, absorption, etc. There are many different approaches for rendering images from these volume datasets. Volume rendering includes only those techniques in which the image generation does not use any intermediate surface representations of the sampled data. These techniques fall into four groups: object-order, image-order, and hybrid algorithms as well as domain methods.l ingmar&s.sunysb.edu, Permission IO runke dij$nl/Jlnrd copies ofnll or part of this tnnterinl for pursosnl or clnssroom 11s~ is gmakd without Re provided thnt tke copies nre not mnde or disM.Wcd hr prolil or conunercinl advantage, the copyright nolice, ~lle title of he publication nad ils date appar, and nolice is given tl~nt copyriSllt is by penkion ofthe ACM. Inc. To copy otherwise, IO republish, IO post on servr'rs or to redislriibute to lists, requires specific penuirsion and/or Ii'e 1997 SlG(iR.4 PH/Euqtwphics Workhop CopyrigM I937 ACM 0.89791.36l-0/37/S..%3.50Object-order algorithms compute each voxel contribution to all affected image pixels in object-storage order (e.g., splat&g [18]). Image-order algorithms compute each pixel color in image scan line order. For each pixel, a ray is cast into the volume, and all the voxels in the neighborhood of the ray are processed and composited to generate a color value for the pixel (e.g., ray-casting [7], volumetric raytracing [15]). These algorithms usually deliver higher image quality than object-order algorithms, but they also require more computation and tend to access each voxel multiple times.Hybrid algorithms combine object storage-order voxel access with image scan line processing order (e.g., slice transformation [1], cell-by-cell processing [16], template-based volume viewing [19], shear warp [6]). Good hybrid algorithms combine the high image quality of image-order algorithms with the efficient data acces...

show abstract

“…Many designs have been proposed [2,4,7,8,9,10,11,12,15,16,171, and n few either simulated or actually built [l, 4, 9, 10, 151. Volume visualization is compute intensive and very demanding on system resources like CPU compute power, memory bandwidth and bus bandwidth.…”

Section: Introductionmentioning

confidence: 99%

Design of a high performance volume visualization system

Lichtenbelt¹

1997

Proceedings of the ACM SIGGRAPH/EUROGRAPHICS Workshop on Graphics Hardware - HWWS '97

View full text Add to dashboard Cite

Visunlizing three dimensional discrete datasets has been a topic of many research projects and papers in the past decade. We discuss the issues that come up when designing a whole computer system cnpable of visualizing these datasets in real time. We explain the three way chicken and egg problem and discuss HewlettPackard's effort at breaking it with the Voxelator API extensions to OpenGL. We enumerate what a good hardware design should accomplish, We discuss what system issues are important and show how to integrate volume visualization hardware in one of Hewlett-Packard's graphics accelerators, the VISUALIZE-48XP. We show why the Voxelator is an efficient and well designed API by explnining how various existing hardware engines will easily tit into the Voxelator framework.

show abstract

Implementations of Cube-4 on the Teramac custom computing machine

Cited by 9 publications

References 5 publications

Real-time volume rendering

Real-time volume rendering

A ray-slice-sweep volume rendering engine

Design of a high performance volume visualization system

Contact Info

Product

Resources

About