Alark Joshi

Art-inspired Techniques for Visualizing Time-varying Data: Novel Techniques for Effectively Visualizingtime-varying Data Alark Joshi

Time-varying data is generated or measured in domainsas diverse as computational fluid dynamics, hurricanesand so on. The data generally contains specificfeatures as well as events that are of interest to anapplicationexpert. We present novel art-inspired techniques forvisualizing time-varying data that conveyschange over time in a single image. Illustrators andpainters have conveyed concepts such as motion and other change over time using illustrative cues. Time-varying data in which three-dimensionalfeatures are moving over time, can be effectivelyvisualized using our illustrative techniquessuch as speedlines, flow ribbons, strobe silhouettesand opacity-based techniques. We appliedthese techniques to computational fluid dynamics dataas well as hurricanes. Our illustration-inspiredtechniquesallowed our collaborators to obtain further insightinto the evolution and intensification of a hurricane. For effectively visualizing time-varying data such asinfant mortality, global rainfall, voting patternsover time, we developed a novel pointillism-based techniqueinspired by Seurat. These techniques convey variability as well as trends over time in a single image.