Coherent Visualization of 2D Scalar Field Contour Ensembles With Probabilistic Latent Space Modeling

2026-07-27

AI Summary

This work addresses the inconsistency among different visual representations—such as contour boxplots and density plots—in ensembles of 2D scalar fields by proposing a unified visualization framework grounded in a probabilistic latent space. The approach employs a variational autoencoder to map ensemble members into this latent space, where data depth is computed using pairwise similarity matrices and integrated with uncertainty-aware clustering. Leveraging the probabilistic distribution in the latent space, the method generates density plots that better preserve the statistical characteristics of the original ensemble. By uniquely combining probabilistic latent space modeling with data depth and clustering, this framework achieves substantially improved consistency across multiple visual views and superior visualization quality compared to existing techniques, as demonstrated on both synthetic and real-world datasets.

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.

Problem

• contour ensembles
• coherent visualization
• 2D scalar field
• uncertainty visualization
• ensemble visualization

Innovation

• probabilistic latent space
• variational autoencoder
• contour ensemble visualization
• uncertainty-aware clustering
• data depth