Graphics Reference
In-Depth Information
3.5 Conclusion
We have presented a novel implementation for quadtrees running completely on
the GPU. It takes advantage of linear trees to alleviate the recursive nature of
common tree implementations, offering a simple and ecient data structure for
parallel processors. While this work focuses on quadtrees, this representation
can also be used for higher dimensional trees, such as octrees. Using the same
distance-based criterion, the
algorithm [Lengyel 10] could be employed
to produce a crack-free volume extractor, running entirely on the GPU. We expect
such an implementation to be extremely fast as well.
Transvoxel l
Acknowledgements
This work was partly funded in Canada by GRAND and in France by ANR.
Jonathan Dupuy acknowledges additional financial support from an Explo'ra Doc
grant from “region Rhone-Alpes,” France.
Bibliography
[AMD 10] AMD. “Tessellation for All.” http://community.amd.com/
[Andersson 07] Johan Andersson. “Terrain Rendering in Frostbite Using Proce-
dural Shader Splatting.” In
ACM SIGGRAPH 2007 Courses
, pp. 38-58.
New York: ACM, 2007.
[Cantlay 11] Iain Cantlay. “DirectX 11 Terrain Tessellation.” White pa-
sdk/11/TerrainTessellation WhitePaper.pdf, 2011.
[Fernandes and Oliveira 12] Antonio Ramires Fernandes and Bruno Oliveira.
“GPU Tessellation: We Still Have a LOD of Terrain to Cover.” In
OpenGL
Insights
, edited by Patrick Cozzi and Christophe Riccio, pp. 145-162. Boca
Raton, FL: CRC Press, 2012.
[Gargantini 82] Irene Gargantini. “An Effective Way to Represent Quadtrees.”
Communications of the ACM
25:12 (1982), 905-910.
[Gateau 07] Samuel Gateau. “Solid Wireframe.” http://developer.
SolidWireframe.pdf, 2007.
[Lengyel 10] Eric Lengyel. “Voxel-Based Terrain for Real-Time Virtual Simula-
tions.” Ph.D. thesis, University of California, Davis, CA, 2010.
