A survey of stereoscopic visualization support in mainstream geographic information systems

Abstract

Principles of stereoscopic vision have long been used in geographic data processing and visualization, e.g. Laussedat created what he called metrophotography in 1895 [2][3] and a stereoautograph for plotting from terrestrial photographs was produced in 1911 [2]. Today, a common and established use of stereoscopy in geographic domain is three-dimensional (3D) modeling of terrains and cities based on aerial, satellite and/or terrestrial imagery with photogrammetric techniques. In these systems, commonly, stereoscopic imagery is used as an input and a vector 3D representation of the terrain or city is obtained as an output. Stereoscopy has also been used for close-range photogrammetry, however within the scope of this paper the focus is on the cartographic use of stereoscopic visualizations. In recent years, a number of mainstream digital devices with stereoscopic capabilities are marketed (e.g. [4], [5], [6]), which indicates that we can expect to see more stereoscopic content in the near future [1]. In anticipation of such a development, it is also plausible to expect that stereoscopic geo-data will become more available for everyday use. It is, therefore, interesting to document which stereoscopic visualization features are supported by mainstream geographic information systems (GIS) software and whether these features can be improved. This work presents a survey on current stereoscopic visualization support in geographic information (GI) systems. The underlying objective is to outline future research needs by examining current stereoscopic 3D visualization techniques and tools in current GISs. This paper focuses on visualization modules of GIS software, and tries to answer the question “How commonly is stereoscopic visualization supported in GIS? To achieve this, first whether such support exits or not in GIS software is documented, then supported types of stereoscopic operations and viewing methods are reported. Motivated by improving efficiency of stereoscopic visualizations in cartographic domain, within this survey, it is also examined whether these software offer level of detail (LOD) management support for stereoscopic datasets. Findings from the study are meant as a guide to researchers and practitioners to assess the current state of True 3D support in widely used GIS software.