Information

Replaced by Rautek-2007-SLI.

Abstract

Direct volume rendering techniques map volumetric attributes (e.g. density, gradient magnitude, etc.) to visual styles. Commonly this mapping is specified by a transfer function. The specification of transfer functions is a complex task and requires expert knowledge about the underlying rendering technique. In the case of multiple volumetric attributes and multiple visual styles the specification of the multi-dimensional transfer function becomes more challenging and non-intuitive. We present a novel methodology for the specification of a mapping from several volumetric attributes to multiple illustrative visual styles. We introduce semantic layers that allow a domain expert to specify the mapping in the natural language of the domain. A semantic layer defines the mapping of volumetric attributes to one visual style. Volumetric attributes and visual styles are represented as fuzzy sets. The mapping is specified by rules that are evaluated with fuzzy logic arithmetics. The user specifies the fuzzy sets and the rules without special knowledge about the underlying rendering technique. Semantic layers allow for a linguistic specification of the mapping from attributes to visual styles replacing the traditional transfer function specification.

Additional Files and Images

Additional files:
technical report
technical report		 video (divx)
video (divx)		 video (mpeg4)
video (mpeg4)


BibTeX

Download BibTeX-Entry
@techreport{TR-186-2-07-05,
  title =      "Semantic Layers for Illustrative Volume Rendering",
  author =     "Peter Rautek and Stefan Bruckner and Meister Eduard
               Gr{\"o}ller",
  year =       "2007",
  abstract =   "Direct volume rendering techniques map volumetric attributes
               (e.g. density, gradient magnitude, etc.) to visual styles.
               Commonly this mapping is specified by a transfer function.
               The specification of transfer functions is a complex task
               and requires expert knowledge about the underlying rendering
               technique. In the case of multiple volumetric attributes and
               multiple visual styles the specification of the
               multi-dimensional transfer function becomes more challenging
               and non-intuitive. We present a novel methodology for the
               specification of a mapping from several volumetric
               attributes to multiple illustrative visual styles.  We
               introduce semantic layers that allow a domain expert to
               specify the mapping in the natural language of the domain. 
               A semantic layer defines the mapping of volumetric
               attributes to one visual style.  Volumetric attributes and
               visual styles are represented as fuzzy sets. The mapping is
               specified by rules that are evaluated with fuzzy logic
               arithmetics. The user specifies the fuzzy sets and the rules
               without special knowledge about the underlying rendering
               technique. Semantic layers allow for a linguistic
               specification of the mapping from attributes to visual
               styles replacing the traditional transfer function
               specification. ",
  address =    "Favoritenstrasse 9-11/186, A-1040 Vienna, Austria",
  institution = "Institute of Computer Graphics and Algorithms, Vienna
               University of Technology",
  note =       "human contact: technical-report@cg.tuwien.ac.at",
  month =      apr,
  number =     "TR-186-2-07-05",
  keywords =   "Illustrative Visualization, Focus+Context Techniques, Volume
               Visualization",
  URL =        "http://www.cg.tuwien.ac.at/research/publications/2007/TR-186-2-07-05/",
}