ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume V-2-2022
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2022, 317–324, 2022
https://doi.org/10.5194/isprs-annals-V-2-2022-317-2022
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2022, 317–324, 2022
https://doi.org/10.5194/isprs-annals-V-2-2022-317-2022
 
17 May 2022
17 May 2022

AUTOMATIC TRAINING DATA GENERATION IN DEEP LEARNING-AIDED SEMANTIC SEGMENTATION OF HERITAGE BUILDINGS

A. Murtiyoso1, F. Matrone2, M. Martini3, A. Lingua2, P. Grussenmeyer4, and R. Pierdicca5 A. Murtiyoso et al.
  • 1Forest Resources Management Group, Institute of Terrestrial Ecosystems, Department of Environmental Systems Science, ETH Zürich, Switzerland
  • 2Department of Environment, Land and Infrastructure Engineering, Politecnico di Torino, Torino, Italy
  • 3Department of Information Engineering, Università Politecnica delle Marche, Ancona, Italy
  • 4Université de Strasbourg, CNRS, INSA Strasbourg, ICube Laboratory UMR 7357, Photogrammetry and Geomatics Group, Strasbourg, France
  • 5Department of Civil and Building Engineering and Architecture, Università Politecnica delle Marche, Ancona, Italy

Keywords: Training data, Automation, Deep learning, Point cloud, Heritage, Semantic segmentation

Abstract. In the geomatics domain the use of deep learning, a subset of machine learning, is becoming more and more widespread. In this context, the 3D semantic segmentation of heritage point clouds presents an interesting and promising approach for modelling automation, in light of the heterogeneous nature of historical building styles and features. However, this heterogeneity also presents an obstacle in terms of generating the training data for use in deep learning, hitherto performed largely manually. The current generally low availability of labelled data also presents a motivation to aid the process of training data generation. In this paper, we propose the use of approaches based on geometric rules to automate to a certain degree this task. One object class will be discussed in this paper, namely the pillars class. Results show that the approach managed to extract pillars with satisfactory quality (98.5% of correctly detected pillars with the proposed algorithm). Tests were also performed to use the outputs in a deep learning segmentation setting, with a favourable outcome in terms of reducing the overall labelling time (−66.5%). Certain particularities were nevertheless observed, which also influence the result of the deep learning segmentation.