DocumentCode :
2413810
Title :
Example-based volume illustrations
Author :
Lu, Albert ; Ebert, David S.
Author_Institution :
Purdue Univ., West Lafayette, IN, USA
fYear :
2005
fDate :
23-28 Oct. 2005
Firstpage :
655
Lastpage :
662
Abstract :
Scientific illustrations use accepted conventions and methodologies to effectively convey object properties and improve our understanding. We present a method to illustrate volume datasets by emulating example illustrations. As with technical illustrations, our volume illustrations more clearly delineate objects, enrich details, and artistically visualize volume datasets. For both color and scalar 3D volumes, we have developed an automatic color transfer method based on the clustering and similarities in the example illustrations and volume sources. As an extension to 2D Wang tiles, we provide a new, general texture synthesis method for Wang cubes that solves the edge discontinuity problem. We have developed a 2D illustrative slice viewer and a GPU-based direct volume rendering system that uses these non-periodic 3D textures to generate illustrative results similar to the 2D examples. Both applications simulate scientific illustrations to provide more information than the original data and visualize objects more effectively, while only requiring simple user interaction.
Keywords :
colour graphics; data visualisation; image texture; rendering (computer graphics); 2D Wang tiles; 2D illustrative slice viewer; GPU-based direct volume rendering system; Wang cubes; automatic color transfer method; nonperiodic 3D textures; scientific illustrations; texture synthesis method; volume datasets illustrations; Application software; Biomedical imaging; Computational modeling; Computer graphics; Humans; Magnetic resonance imaging; Rendering (computer graphics); Shape; Tiles; Visualization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Visualization, 2005. VIS 05. IEEE
Print_ISBN :
0-7803-9462-3
Type :
conf
DOI :
10.1109/VISUAL.2005.1532854
Filename :
1532854
Link To Document :
بازگشت