Visual Analysis of Large Multivariate Scattered Data using Clustering and Probabilistic Summaries

Rapidly growing data sizes of scientific simulations pose significant challenges for interactive visualization and analysis techniques. In this work, we propose a compact probabilistic representation to interactively visualize large scattered datasets. contrast previous approaches that represent blocks volumetric using probability distributions, model clusters arbitrarily structured multivariate data. detail, discuss how efficiently store high-dimensional distribution each cluster. We observe it suffices consider low-dimensional marginal distributions two or three dimensions at time employ common visual Based on observation, by combinations Gaussian mixture models. the application techniques representation. particular, investigate several frequency-based views, such as density plots in 1D 2D, density-based parallel coordinates, histogram. uncertainty introduced representation, level-of-detail mechanism, explicitly outliers. Furthermore, spatial splatting anisotropic 3D Gaussians which derive closed-form solution. Lastly, describe brushing linking clustered Our evaluation large, real-world datasets demonstrates scaling our approach.