Dynamic View Selection

for Time-varying Volumes

Abstract

Publication

Animation is an effective way to show how time-varying phenomena evolve over time. A key issue of generating a good animation is to select ideal views through which the user can perceive the maximum amount of information from the time-varying dataset.

In this paper, we rst propose an improved view selection method for static data. The method measures the quality of a static view by analyzing the opacity, color and curvature distributions of the corresponding volume rendering images from the given view. Our view selection metric prefers an even opacity distribution with a larger projection area, a larger area of salient features' colors with an even distribution among the salient features, and more perceived curvatures. We use this static view selection method and a dynamic programming approach to select time-varying views. The time-varying view selection

maximizes the information perceived from the time-varying dataset based on the constraints that the time-varying view should show smooth changes of direction and near-constant speed.

We also introduce a method that allows the user to generate a smooth transition between any two views in a given time step, with the perceived information maximized as well. By combining the static and dynamic view selection methods, the users are able to generate a time-varying view that shows the maximum amount of information from a time-varying data set.

PDF  

Download