Screen-space silhouettes for visualizing ensembles of 3D isosurfaces

Abstract

Figure 1: Visualization of a weather forecast ensemble from the ECMWF Prediction System [29]. This ensemble comprises wind velocity data and consists of 50 members. For one isovalue, all members are visualized as silhouettes of isosurfaces. Additionally, a mean ensemble member is rendered as a gray isosurface to enhance the visual perception of the spatial context. Color is used to cluster members by their similarity. ABSTRACTVisualizing sets of isosurfaces from 3D scalar ensemble fields is a difficult task due to inherent occlusion effects, yet it is often required to analyze the uncertainty represented by such an ensemble. In this paper, we present a novel visualization technique for ensembles of isosurfaces based on screen-space silhouettes. By using silhouettes, the displayed information is reduced to avoid occlusions, yet the major shape of the surfaces can be maintained. Our approach preserves spatial coherence and does not make any assumption about the underlying surface distribution. By providing additional mechanisms, i.e., picking, clustering, cutting and animation, we enable the user to explore an ensemble of surfaces interactively.

eBoF: Interactive Temporal Correlation Analysis for Ensemble Data Based on Bag-of-Features

“…For example, Ferstl et al 12 use PCA to convert streamlines into a structure preserving Euclidean space. Furthermore, a lot of approaches are implemented by feature extraction algorithms like k-series in the work 5,36,6 , and the work based on DBSCAN 10,26 , AHC 12,13,37,38,39,40 and topic modeling 41,42,43,44,45,46,47,48,49,50,51,52,53 .…”

Visualization of WiFi Signals Using Programmable Transfer Functions

“…This method produces a bubble-like shading effect. This effect is described in Demir et al [30]. Programmable Transfer Functions can not only implement the silhouette shading effect; they also enable the user to manipulate the silhouette parameters in real time.…”

Making Sense of Scientific Simulation Ensembles With Semantic Interaction

“…Increasing in complexity, we see the use of pairwise sequential animation and screen door tinting to show the differences between ensemble members using value changes to field points [PPA*12], the comparison of 2D function ensembles on three levels of details (i.e. surface plot, domain‐oriented and member‐oriented) [PPBT12], the comparison of 3D scalar field ensembles using mean isosurface [DKW16] and grouping ensemble members using hierarchical clustering [HHBY16]. Although these tools offer different ways to visualize and compare different data types, they are limited in the number of ensemble members that they visualize.…”