To be effective visual representations must be adapted to their respective context of use, especially in so-called Smart Visual Interfaces striving to present specifically those information required for the task at hand. This thesis proposes a generic approach that facilitate the automatic generation of task-specific visual representations from suitable task descriptions. It is discussed how the approach is applied to four principal content types raster images, 2D vector and 3D graphics as well as data visualizations, and how existing display techniques can be integrated into the approach.
