Istituto di Scienza e Tecnologie dell'Informazione     
Spinelli A., Ganovelli F., Montani C., Scopigno R. Recovering 3D architectural information from dense digital models of buildings. In: Eurographics Italian Chapter 2006 (Catania, Italy, 22-24 February 2006). Atti, pp. 177 - 182. Fellner, Dieter (eds.). Eurographics Association, 2006.
In recent years the progress of 3D scanning technologies and the consequent growing commercialization of scanners opened a large spectrum of opportunities for many professionals. In particular, architects and engineers may access to a digital model of a building without having to model it using a CAD software. On the other hand, there are two important differences between the digitized and the handcrafted model. The first is the absence of interpretation. The digitized model is only a set of polygons that describe, possibly in a very accurate manner, the scanned object. It does not provide the user with any other information about what a surface is (a wall, a window, an arch etc.) that, conversely, can be incorporated during the editing in a CAD session. The second difference is excess of realism. In the digitized models the are no planar walls, no right angles, no straight edges, simply because they are not, at the millimetric scale. Unfortunately, if a model must be used in a FEM simulation, for example, a CAD like model would be required. This paper describes an application framework and some techniques that have been implemented to help a non computer-graphics user in handling digital models of buildings acquired using 3D scanning. The techniques permit to visualize efficiently the models independently from their size, recover 3D information (measurements, sections), extract geometric features and fit high level geometric primitives.
URL: http://vcg.isti.cnr.it/Publications/2006/SGMS06/paper.pdf
Subject 3D scanning
Feature extraction
Data fitting
I.4.8 Scene Analysis. Range data

Icona documento 1) Download Document PDF

Icona documento Open access Icona documento Restricted Icona documento Private


Per ulteriori informazioni, contattare: Librarian http://puma.isti.cnr.it

Valid HTML 4.0 Transitional