An intuitive Sketch-based Transfer Function Design via Contextual and Regional Labelling

Abstract

Transfer function (TF) in direct volume rendering serves to identify and emphasize features of interest (FOIs) and their contextual and regional information for improved visualization. Conventional TF design is not intuitive and usually a 'trial-and-error' process for most users. In an intensity-based one-dimensional (1D) histogram TF, for example, a user needs to repetitively adjust intensity ranges (to identify FOIs) and then assign color and opacity values to the selected range (to emphasize FOIs). In this paper, we propose an intuitive sketch-based interaction technique to design TFs. Our technique enables the user to identify FOIs along the user's viewing ray, with the aid of contextual and regional labels automatically derived from two-dimensional (2D) image slices reconstructed from the ray. For FOI identification, the user makes a sketch on the 2D image slice. Our technique automatically generates an intensity-based 1D TF where the opacity and color values of the intensity range for the FOIs are derived according to their distance from the user's viewpoint and this allows all FOIs along the ray to be visible at once. We show the capabilities of our technique with visualizations on different volumetric data sets, and highlight its advantages when compared to the conventional histogram TF design.