Persistence Atlas for Critical Point Variability in Ensembles

Abstract

Fig. 1. Persistence atlas for an ensemble of 45 von Kármán vortex streets (scalar data: orthogonal component of the curl). (a) Critical points (minima and maxima, scaled by persistence) of a few representative ensemble members (one color per member) exhibit clearly distinct layout patterns in terms of position and number of vortices, revealing high spatial and trend variabilities within the ensemble. (b) Mandatory critical points (minimal regions where at least one critical point is guaranteed to occur for every member of the ensemble) are thus particularly conservative given these variabilities and identify only one region per side of the vortex street (blue: minimum, green: maximum). (c) The persistence atlas addresses this issue by analyzing the structure of the ensemble in terms of critical point layouts and provides low dimensional embeddings of the members where statistical tasks, such as clustering, can be easily carried out. In particular, our approach automatically identified five clusters, (d) to (h), corresponding to five distinct trends in critical point layouts (five viscosity regimes). Per cluster mandatory critical points provide more accurate and useful critical point predictions (colored regions, (d) to (h)), revealing an increasing number of vortices and a decreasing spatial variability for increasing Reynolds numbers (left to right). The background color map shows the mean scalar field for the entire ensemble, (a) and (b), and individual clusters, (d) to (h).Abstract-This paper presents a new approach for the visualization and analysis of the spatial variability of features of interest represented by critical points in ensemble data. Our framework, called Persistence Atlas, enables the visualization of the dominant spatial patterns of critical points, along with statistics regarding their occurrence in the ensemble. The persistence atlas represents in the geometrical domain each dominant pattern in the form of a confidence map for the appearance of critical points. As a by-product, our method also provides 2-dimensional layouts of the entire ensemble, highlighting the main trends at a global level. Our approach is based on the new notion of Persistence Map, a measure of the geometrical density in critical points which leverages the robustness to noise of topological persistence to better emphasize salient features. We show how to leverage spectral embedding to represent the ensemble members as points in a low-dimensional Euclidean space, where distances between points measure the dissimilarities between critical point layouts and where statistical tasks, such as clustering, can be easily carried out. Further, we show how the notion of mandatory critical point can be leveraged to evaluate for each cluster confidence regions for the appearance of critical points. Most of the steps of this framework can be trivially parallelized and we show how to efficiently implement them. Extensive experiments demonstrate the relevance of our approach. The accuracy of the confidence regions provided by the per...

Progressive Wasserstein Barycenters of Persistence Diagrams

“…Overall, our approach provides the same clustering results than Favelier et al [32]: the returned clusterings are correct for both approaches, for all of the above data sets. However, once the input persistence diagrams are available, our algorithm computes within a time constraint of ten seconds only, while the approach by Favelier et al requires up to hundreds of seconds (on the same hardware) to compute intermediate representations (Persistence Maps) which are not needed in our work.…”

“…Our experiments were performed on a variety of simulated and acquired 2D and 3D ensembles, taken from Favelier et al [32]. The Gaussians ensemble contains 100 2D synthetic noisy members, with 3 patterns of Gaussians (Fig.…”

State of the Art in Time‐Dependent Flow Topology: Interpreting Physical Meaningfulness Through Mathematical Properties

“…Favelier et al . [FFST19] visualize positional uncertainties of critical points using persistence‐based clustering.…”

Visualizing the Stability of 2D Point Sets from Dimensionality Reduction Techniques

“…Alternatively, clustering has been used to group ensemble members regarding similar data characteristics [BM10, TN14, OLK*14, FBW16, FFST19]. While these techniques compare ensemble members to each other, our approach aims at finding groups of elements in each member which remain ‘close’ to each other in all members.…”