These animations illustrate the operation of the Evita system for a two-dimensional scalar-valued dataset on a rectangular grid. The dataset depicts a simulated turbulent region in the Pacific Ocean that was obtained from an analysis of the velocity field. The dataset contains the scalar swirl parameter, a scalar value that measures bulk rotation of fluid with a "large" value of the parameter (white in the animations) indicating a highly rotational region. The dataset has been segmented into 3 regions-of-interest (ROIs) plus background.
During the visualization session, the system allows the user to select ROIs, and selected ROIS are given increased priority in the visualization process, i.e., information from selected ROIs is sent before the remainder of the dataset. In the compressed representation of the dataset produced by the Evita system preprocessor, each ROI is stored as a series of "enhancement layers", with each layer offering progressive refinement of the visualization of the ROI. During the visualization session, the enhancement layers are accessed in the order that the user client determines is the most effective for data exploration, and this ordering can be changed dynamically as the visualization progresses.
- In this animation, one layer of enhancement is visualized for each ROI. The user then selects ROI 2. At this point, the system transmits all remaining enhancement layers for ROI 2, after which, transmission continues for the other ROIs and the background. We note that other, more general, types of priority scheduling are possible; e.g., the system could merely increase the speed that information about the selected ROI is sent rather than devoting all transmission to the selected ROI as is illustrated in this animation.