Direct volume rendering by transfer function morphing

Abstract

Direct volume rendering has been widely used by physicians, scientists, and engineers in a variety of applications. In a direct volume rendering process, transfer functions are used to specify the relationships between volume data values and optical properties in order to reveal inner structures in a volume. However, trial-and-error is one of the widely used approaches for finding good transfer functions to visualize volume data. In this paper, we propose a new approach to automatically visualize volume data by transfer function morphing. With our approach, the user can just define the start transfer functions and end transfer functions for a volume, and all the intermediate transfer functions are generated based on a morphing factor function (linear interpolation is used as an example morphing factor function in our study). These transfer functions are used for visualizing the volume simultaneously as they are generated. Experimental results on various volume datasets show that our approach is useful for the user to get the insight of volume data effectively and efficiently.