Coherent Visualization of 2D Scalar Field Contour Ensembles With Probabilistic Latent Space Modeling
2026-07-27 • Graphics
Graphics
AI summaryⓘ
The authors created a new way to visualize groups of 2D contour lines generated from sets of data called ensembles. They use a machine learning tool called a variational autoencoder (VAE) to represent each data member in a simpler form, which helps them measure similarities and group the data effectively. This approach makes it easier to show uncertainty and creates clearer visual summaries, like density plots, that better match the original data than older methods. They tested their method with both made-up and real-world data to show it works well.
contour ensemblesvariational autoencoderlatent spaceprobabilistic modelingdata depthuncertainty-aware clusteringdensity plots2D scalar fieldsvisualization
Authors
Cenyang Wu, Runhao Lin, Qinhan Yu, Liang Zhou
Abstract
We present a new visualization method for contour ensembles through probabilistic modeling. We aim to improve the coherence between different visual representations, such as contour boxplots and density plots for a 2D scalar field ensemble. We model each ensemble member with a probabilistic representation in the latent space, i.e., a lower-dimensional representation of spatial data features, of a variational autoencoder (VAE). Thereafter, efficient data depth computation and uncertainty-aware clustering are supported based on a matrix of pair-wise similarity measurements of members. We estimate the underlying probability distribution by leveraging the power of VAE to create density plots that align more coherently with member distributions than existing methods. The effectiveness of our method is evaluated through numerical comparisons with existing techniques, and visualization examples of synthetic and real-world ensemble datasets.