The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B2-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2022, 77–84, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-77-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIII-B2-2022, 77–84, 2022
https://doi.org/10.5194/isprs-archives-XLIII-B2-2022-77-2022
 
30 May 2022
30 May 2022

AERIAL TRIANGULATION WITH LEARNING-BASED TIE POINTS

F. Remondino1, L. Morelli1,2, E. Stathopoulou1,3, M. Elhashash4,6, and R. Qin4,5,6,7 F. Remondino et al.
  • 13D Optical Metrology (3DOM) unit, Bruno Kessler Foundation (FBK), Trento, Italy
  • 2Dept. of Civil, Environmental and Mechanical Engineering (DICAM), University of Trento, Italy
  • 3Laboratory of Photogrammetry, National Technical University of Athens (NTUA), Athens, Greece
  • 4Geospatial Data Analytics Lab, The Ohio State University, Columbus, USA
  • 5Department of Civil, Environmental and Geodetic Engineering, The Ohio State University, Columbus, USA
  • 6Department of Electrical and Computer Engineering, The Ohio State University, Columbus, USA
  • 7Translational Data Analytics Institute, The Ohio State University, Columbus, USA

Keywords: aerial triangulation, tie points, deep learning, image matching, accuracy analysis

Abstract. Aerial triangulation (AT) has reached outstanding progress in the last decades, and now fully automated solutions for nadir and oblique images are available. Usually, image correspondences (tie points) are found using hand-crafted methods, such as SIFT or its variants. But in the last years, there were many investigations and developments to promote the use of machine and deep learning solutions within the photogrammetric processing pipeline. The paper explores learning-based methods for the extraction of tie points in aerial image blocks. Image correspondences are used to perform aerial triangulation (AT) and successively generate dense point clouds. Two different datasets are used to compare conventional hand-crafted detector/descriptor methods with respect to learning-based methods. Accuracy analyses are performed using GCPs as well as ground truth LiDAR point clouds. Results confirm the potential of learningbased methods in finding reliable image correspondences in the aerial block, still showing space for improvements due to camera rotations.