Visualizing Spatiotemporal Models with Virtual Reality: from Fully Immersive Environments to Applications in Stereoscopic View. (15th January 2019)

• Record Type: Journal Article
• Authors: Castruccio, Stefano; Genton, Marc G.; Sun, Ying
• Is Part Of: Journal of the Royal Statistical Society. Volume 182:Number 2(2019)
• Page Start: 379
• Page End: 387
• Publication Date: 2019-01-15
• Subjects: Data visualization -- Global models -- Spatiotemporal statistics -- Virtual reality; Social sciences -- Statistical methods -- Periodicals; Statistics -- Periodicals; 300.15195
• DOI: 10.1111/rssa.12381
• Languages: English
• ISSNs: 0964-1998

Abstract

Summary: Recent advances in computing hardware and software present an unprecedented opportunity for statisticians who work with data indexed in space and time to visualize, explore and assess the structure of the data and to improve resulting statistical models. We present results of a 3-year collaboration with a team of visualization experts on the use of stereoscopic view and virtual reality (VR) to visualize spatiotemporal data with animations on non-trivial manifolds. We first present our experience with fully immersive VR with motion tracking devices that enable users to explore global three-dimensional time–temperature fields on a spherical shell interactively. We then introduce a suite of applications with VR mode, freely available for smartphones, to port a visualization experience to any interested people. We also discuss recent work with head-mounted devices such as a VR headset with motion tracking sensors.