Project A01

A01 | Uncertainty Quantification and Analysis in Visualization

Prof. Daniel Weiskopf, University of Stuttgart
Email | Website

Prof. Oliver Deussen, University of Konstanz
Email | Website

Prof. Andreas Bulling, University of Stuttgart
Email | Website

Prof. Miriam Butt, University of Konstanz
Email | Website

Dr. Marina Evers, University of Stuttgart – Email | Website

David Hägele, University of Stuttgart – Email | Website

Dr. Rebecca Kehlbeck, University of Konstanz – Email | Website

Patrick Paetzold, University of Konstanz – Email | Website

Michael Stroh, University of Konstanz – Email | Website

Sita Vriend, University of Stuttgart - Email | Website

Yumeng Xue, University of Konstanz, Email | Website

About this project
Results

Our long-term goal is the modeling, handling, and quantification of uncertainty throughout the complete visual computing process, from visual-computing specific models of uncertainty to the visual representation of, and interaction with, uncertainty information. We aim to investigate models and methods for quantifying individual sources of uncertainty, the propagation of uncertainty through the visual computing pipeline, and the impact of uncertainty on the components of the visual computing process. Furthermore, we will advance the visualization and visual analytics of uncertainty, and investigate several different application examples.

Research Questions

How can we integrate and combine the different aspects of uncertainty in the various subareas of visual computing?

How can uncertainty be modeled and propagated through the different steps of the visualization process?

What are effective ways to communicate uncertainty to humans and how can they meaningfully interact with such displays?

Which methods are best suited to evaluate uncertainty visualization, and how can we quantify human understanding of uncertainty?

Uncertainty-Aware Principal Component Analysis. The data consists of multivariate probability distributions, the plots shows the impact of different levels of variance.

Fig. 1: Uncertainty-Aware Principal Component Analysis. The data consists of multivariate probability distributions, the plots shows the impact of different levels of variance.

Bubble Treemap for visualizing uncertainty in hierarchical data. The uncertainty of the values in the hierarchy is visually encoded by the waviness of the outlines.

Fig. 2: Bubble Treemap for visualizing uncertainty in hierarchical data. The uncertainty of the values in the hierarchy is visually encoded by the waviness of the outlines.