Adaptive Hierarchical Visibility in a Tiled Architecture

dc.contributor.authorXie, Fengen_US
dc.contributor.authorShantz, Michaelen_US
dc.contributor.editorA. Kaufmann and W. Strasser and S. Molnar and B.- O. Schneideren_US
dc.date.accessioned2014-02-06T15:04:37Z
dc.date.available2014-02-06T15:04:37Z
dc.date.issued1999en_US
dc.description.abstractThis paper describes a method for occlusion culling in a tiled 3D graphics hardware architecture. Adaptive hierarchical visibility (AHV) is a simplified method for occlusion culling that is integrated into a tiled architecture for hardware rendering. AI-IV constructs a list of polygon bins for each tile where the bins are bucket sorted in order of increasing depth or Z. Polygon bins are rendered starting with the bin closest to the viewer. After some number of bins are rendered, a one layer, hierarchical Zbuffer (HZ) is constructed from the Z-buffer thus far accumulated for the rendered bins. Subsequent bins are rendered by first testing their polygons against the HZ to see if they are hidden. AHV is far simpler to implement in hardware and gives performance that matches or surpasses progressive hierarchical visibility (PHV) methods which update the HZ for each rendered pixel. Results show that AI-IV is superior on scenes with high depth complexity and small polygons. For tiles of widely ranging statistics, AHV competes surprisingly well with PHV. It offers dramatic performance improvement on low cost hardware for scenes of high depth complexity.en_US
dc.description.seriesinformationSIGGRAPH/Eurographics Workshop on Graphics Hardwareen_US
dc.identifier.isbn1-58113-170-4en_US
dc.identifier.issn1727-3471en_US
dc.identifier.urihttps://doi.org/10.2312/EGGH/EGGH99/075-084en_US
dc.publisherThe Eurographics Associationen_US
dc.subject1.3.1 [Computer Graphics]en_US
dc.subjectHardware Architectureen_US
dc.subject1.3.7 [Computer Graphics] Visible Line/Surface Algorithmsen_US
dc.titleAdaptive Hierarchical Visibility in a Tiled Architectureen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
075-084.pdf
Size:
1.59 MB
Format:
Adobe Portable Document Format