The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLI-B4
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B4, 63–69, 2016
https://doi.org/10.5194/isprs-archives-XLI-B4-63-2016
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLI-B4, 63–69, 2016
https://doi.org/10.5194/isprs-archives-XLI-B4-63-2016

  10 Jun 2016

10 Jun 2016

EVALUATION OF MODEL RECOGNITION FOR GRAMMAR-BASED AUTOMATIC 3D BUILDING MODEL RECONSTRUCTION

Qian Yu, Petra Helmholz, and David Belton Qian Yu et al.
  • Cooperative Research Centre for Spatial Sciences (CRCSI) Department of Spatial Sciences, Curtin University GPO Box U 1987, Perth, WA, 6845, Australia

Keywords: building model, evaluation, reconstruction, 3D, grammar

Abstract. In recent years, 3D city models are in high demand by many public and private organisations, and the steadily growing capacity in both quality and quantity are increasing demand. The quality evaluation of these 3D models is a relevant issue both from the scientific and practical points of view. In this paper, we present a method for the quality evaluation of 3D building models which are reconstructed automatically from terrestrial laser scanning (TLS) data based on an attributed building grammar. The entire evaluation process has been performed in all the three dimensions in terms of completeness and correctness of the reconstruction. Six quality measures are introduced to apply on four datasets of reconstructed building models in order to describe the quality of the automatic reconstruction, and also are assessed on their validity from the evaluation point of view.