HyperLabels

In my second PhD project, I developed a method for intuitive navigation in a multi-scale 3D environment. It leverages my previous work with textual labels integrated in the three-dimensional visualization.

There were several issues we wanted to address. Primarily, we had large 3D molecular models that have an inherent hierarchical organization: proteins build protein complexes, which in turn build viral components. We wanted to allow people to navigate this hierarchical structure in an intuitive way.

The basis of the problem was the need to coordinate occlusion management (a typical problem of exporing 3D models) and camera during the hierarchical navigation. The textual labels from my first project (Labels on Levels) offered an intuitive option for grounding the interaction in a familiar concept: hyperlinks.

We therefore made the labeling algorithm annotate scene structures on only one specific level at a time (instead of mixing levels in one view) and we allowed the labels to be clicked. A click on label triggers a level change, which is processed as a change in occlusion management, camera viewpoint, and new set of labels in the 3D scene.

Besides my supervisors, I had a chance to work with two excellent visualization researchers: Tobias Isenberg and Miriah Meyer. The result was another IEEE TVCG publication:

D. Kouřil, T. Isenberg, B. Kozlíková, M. Meyer, M. E. Gröller, I. Viola, “HyperLabels: Browsing of Dense and Hierarchical Molecular 3D Models”, IEEE Transactions on Visualization and Computer Graphics, Early Access, accepted February 2020. Doi: 10.1109/TVCG.2020.2975583

I presented the paper at the IEEE Vis 2020 conference, virtually: