ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume IV-2/W1
05 Oct 2016
 | 05 Oct 2016


R. Neuville, F. Poux, P. Hallot, and R. Billen

Keywords: 3D geovisualisation, 3D cartography, 3D semiotics, camera management, visual variables, user’s query

Abstract. This paper deals with the viewpoint management in 3D environments considering an allocentric environment. The recent advances in computer sciences and the growing number of affordable remote sensors lead to impressive improvements in the 3D visualisation. Despite some research relating to the analysis of visual variables used in 3D environments, we notice that it lacks a real standardisation of 3D representation rules. In this paper we study the “viewpoint” as being the first considered parameter for a normalised visualisation of 3D data. Unlike in a 2D environment, the viewing direction is not only fixed in a top down direction in 3D. A non-optimal camera location means a poor 3D representation in terms of relayed information. Based on this statement we propose a model based on the analysis of the computational display pixels that determines a viewpoint maximising the relayed information according to one kind of query. We developed an OpenGL prototype working on screen pixels that allows to determine the optimal camera location based on a screen pixels colour algorithm. The viewpoint management constitutes a first step towards a normalised 3D geovisualisation.