Optimization in the Co-Registration of Large Point Clouds for Archaeological Visualization

Dominique E Meyer, Jamie Hodgkins, Fabio Negrino, Christopher E Miller, Caley M Orr, Falko Kuester, Stefano Benazzi, Marco Peresani, Julien Riel-Salvatore, David Strait, Matthias Czechowski, Vera Smirnova, Michael Hess

Abstract


Photogrammetry has been used in Archaeological studies to gather 3D data of sites and artifacts. Improvements in imaging systems, stitching algorithms and computation power has increased point cloud sizes and densities. Open-source and commercial softwares often cannot handle very large point clouds, making it cumbersome and difficult to combine large data sets. We propose a novel workflow to optimize the merging of large point clouds and visualizing them. Sites are often reconstructed at different scales ranging from square kilometers for environments to millimeter details on the ground. Aerial Surveys can be used to create large Digital Elevation Models (DEM) and they can contribute 3D models of buildings and architecture which is not easily accessible. Handheld photogrammetry and LIDAR scanning can be used on the ground to gather high accuracy models of structures, interiors and artifacts. A case study of the Arma Vairana Cave in Liguria, Italy, is presented where nine aerial and ground point clouds were merged into a fully immersive point cloud.