Volume II-5
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5, 97-104, 2014
https://doi.org/10.5194/isprsannals-II-5-97-2014
© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., II-5, 97-104, 2014
https://doi.org/10.5194/isprsannals-II-5-97-2014
© Author(s) 2014. This work is distributed under
the Creative Commons Attribution 3.0 License.

  28 May 2014

28 May 2014

3D modelling of trompe l'oeil decorated vaults using dense matching techniques

F. Chiabrando1, A. Lingua1, F. Noardo1, and A. Spano2 F. Chiabrando et al.
  • 1Politecnico di Torino, Dipartimento di Ingegneria dell'Ambiente, del Territorio e delle Infrastrutture, Corso Duca degli Abruzzi 24, 10129 Torino, Italyo
  • 2Politecnico di Torino, Dipartimento di Architettura e Design, Viale Mattioli 39, 10125 Torino, Italy

Keywords: Image matching, digital photogrammetry, area-based-matching, feature-matching, TLS, 3D models, Cultural Heritage

Abstract. Dense matching techniques, implemented in many commercial and open source software, are useful instruments for carrying out a rapid and detailed analysis of complex objects, including various types of details and surfaces. For this reason these tools were tested in the metric survey of a frescoed ceiling in the hall of honour of a baroque building. The surfaces are covered with trompe-l’oeil paintings which theoretically can give a very good texture to automatic matching algorithms but in this case problems arise when attempting to reconstruct the correct geometry: in fact, in correspondence with the main architectonic painted details, the models present some irregularities, unexpectedly coherent with the painted drawing. The photogrammetric models have been compared with data deriving from a LIDAR survey of the same object, to evaluate the entity of this blunder: some profiles of selected sections have been extracted, verifying the different behaviours of the software tools.