Scalable visual queries for data exploration on large, high-resolution 3D displays

Scalable visual queries for data exploration on large, high-resolution 3D displays

November 12, 2012

As the scale and complexity of data continue to grow at unprecedented rates, scientists are increasingly relying on Large, High-Resolution Displays to visualize and analyze scientific datasets. Recent studies have demonstrated the effectiveness of these displays in supporting cognitively demanding data analysis and sensemaking tasks. While there has been an abundance of research on rendering algorithms for large, high-resolution displays, far less effort has gone into designing interactive visual analytic interfaces to effectively leverage these displays in visual exploration and sensemaking scenarios involving large collections of data. In this paper, we present an interactive visual analytics application for the exploration of large trajectory datasets. Our application utilizes large, high-resolution 3D display environments to simultaneously visualize and juxtapose a large number of trajectories. It also integrates a scalable visual query technique, which can be used to quickly formulate and verify hypotheses, encouraging scientists to contemplate multiple competing theories before drawing conclusions. We evaluate our design within the context of a behavioral ecology case study. We also share our observations from a pilot user study to provide insights on how scientists might utilize large display environments in visual exploration and sensemaking scenarios.

Download

Paper PDF

Reference

Reda, Khairi, Andrew Johnson, Victor Mateevitsi, Catherine Offord, and Jason Leigh. "Scalable visual queries for data exploration on large, high-resolution 3D displays." In 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, pp. 196-205. IEEE, 2012.
Image credits: Lance Long, Electronic Visualization Laboratory