Gaze-directed Flow Map Explosions in AR/VR

Origin-Destination (OD) flow maps are a popular topic in contemporary cartographic and GIS research because they may show interesting spatio-temporal patterns, but are difficult to render legibly. As such, they present a classic cartographic generalization problem where complex data at high detail is available, but generalized or summarized representations of these data are necessary for human visualization and cognition. Several scholars have experimented with 3-dimensional representations of movements using arching flow lines, often rendered in virtual or augmented reality settings to allow the user to physically explore their shapes (Andrienko, Andrienko, Chen, Maciejewski, & Zhao, 2017; Graser, Schmidt, Roth, & Brändle, 2017; Yang et al., 2019).

This project will continue work in this vein by explicitly incorporating cartographic generalization. Flows will be aggregated or otherwise generalized using one or more of a diversity of state-of-the-art methods (e.g., flow combination in the manner of Sankey diagrams, edge bundling, simple aggregation). Then, users of an AR/VR application allowing viewing of the generalized flows will be able to visually query for individual or less-generalized flows by fixing their gaze on a generalized representation.

This master’s project involves developing the software that performs all of these tasks: data generalization, visualization, interactivity affordance, gaze detection, and de-generalization. It requires a student comfortable with basic 3D geometry and computer programming, likely in C# and/or Python, and using the Unity development environment.