A Parallel Pipeline Convolution for Perspective Projection in Real-Time Volume Rendering

DSpace/Manakin Repository

A Parallel Pipeline Convolution for Perspective Projection in Real-Time Volume Rendering

Show simple item record

dc.contributor.author Ogata, Masato
dc.contributor.author Ohkami, Takahide
dc.contributor.author Pfister, Hanspeter
dc.contributor.author Lauer, Hugh C.
dc.contributor.author Dohi, Yasunori
dc.date.accessioned 2010-06-16T13:54:12Z
dc.date.issued 2000
dc.identifier.citation Ogata, Masato, Takahide Ohkami, Hanspeter Pfister, Hugh C. Lauer, and Yasunori Dohi. 2000. A parallel pipeline convolution for perspective projection in real-time volume rendering. Journal of the Institute of Image Information and Television Engineers 54(9): 1339-1348. en_US
dc.identifier.issn 1342-6907 en_US
dc.identifier.uri http://nrs.harvard.edu/urn-3:HUL.InstRepos:4238983
dc.description.abstract This paper describes a convolution with a systolic array structure for perspective projection in real-time volume graphics based on the shear-warp method. In the original method, the further the ray proceeds, the more voxels are required to calculate the convolution. The increase in required voxels makes it difficult to implement the method in a VLSI-oriented architecture. We implement a 3D convolution using three serial 1D convolutions along the X, Y, and Z axes, which reduces the number of calculation units from M3 to 3M, where the convolution is calculated for the M3 area. The number of pipelines for the rays is V2 for V3 voxel datasets. If the hardware of a single pipeline can calculate the V rays, then each of the implemented pipelines is assigned to V theoretical pipelines (for V2 rays). The number of hardware pipelines should be much smaller than V theoretical pipelines in actual implementation. We folded the theoretical pipelines and reduced them to a certain number of hardware pipelines. We examined the relation between the folding process and its necessary time delay. The architecture can generate an image of a 2563 voxel dataset (V=256) at 30Hz with four pipelines. In addition, the architecture can be extended easily for 5123 (V=512) and 10243 (V=1024) datasets, with 32 pipelines and 256 pipelines. Our architecture has processing scalability. en_US
dc.description.sponsorship Engineering and Applied Sciences en_US
dc.language.iso en_US en_US
dc.publisher Dō Gakkai en_US
dc.relation.isversionof http://sciencelinks.jp/j-east/article/200102/000020010200A0952145.php en_US
dc.relation.hasversion http://gvi.seas.harvard.edu/sites/all/files/Ieicee5.pdf en_US
dash.license LAA
dc.title A Parallel Pipeline Convolution for Perspective Projection in Real-Time Volume Rendering en_US
dc.type Journal Article en_US
dc.description.version Accepted Manuscript en_US
dc.relation.journal Journal of the Institute of Image Information and Television Engineers en_US
dash.depositing.author Pfister, Hanspeter
dc.date.available 2010-06-16T13:54:12Z

Files in this item

Files Size Format View
Ogata_Parallel.pdf 336.0Kb PDF View/Open

This item appears in the following Collection(s)

  • FAS Scholarly Articles [7218]
    Peer reviewed scholarly articles from the Faculty of Arts and Sciences of Harvard University

Show simple item record

 
 

Search DASH


Advanced Search
 
 

Submitters