Title: Improving Locality for Progressive Radiosity Algorithm: A Study based on the Blocking Transformation of the Scene
Authors: Sanjurjo, J. R.
Amor, M.
Boo, M.
Doallo, R.
Citation: WSCG '2007: Full Papers Proceedings: The 15th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2007 in co-operation with EUROGRAPHICS: University of West Bohemia Plzen Czech Republic, January 29 – February 1, 2007, p. 113-120.
Issue Date: 2007
Publisher: Václav Skala - UNION Agency
Document type: konferenční příspěvek
URI: http://wscg.zcu.cz/wscg2007/Papers_2007/full/!WSCG2007_Full_Proceedings_Final-1.zip
ISBN: 978-80-86943-98-5
Keywords: globální osvětlení;radiozita;lokalita dat;dělení scény
Keywords in different language: global illumination;radiosity;data locality;scene partition
Abstract: Efficient global illumination is an important challenge in computer graphics. Themain problemof these algorithms is their associated execution time and storage requirements. This is a handicap for simulating large scenes and reducing these costs is a constant research objective of the computer graphics community. In this paper we present a performance analysis of the progressive radiosity algorithm based on the blocking transformation of the scene according to a uniform partition. The data locality exploitation and the associated decrement in cache misses permit the reduction of the execution time of the algorithm. An extended analysis of the influence of the scene subdivision on the execution time requirements is presented. Because of our analysis, we conclude that important performance improvements in terms of execution time are obtained with this technique.
Rights: © Václav Skala - UNION Agency
Appears in Collections:WSCG '2007: Full Papers Proceedings

Files in This Item:
File Description SizeFormat 
Sanjurjo.pdfPlný text961,33 kBAdobe PDFView/Open
Sanjurjo_prezentace.zipPrezentace691,55 kBZIPView/Open

Please use this identifier to cite or link to this item: http://hdl.handle.net/11025/10998

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.