Abstract
Cyber-archaeology represents the marriage of archaeology, computer science, engineering, and the natural sciences with the aim of taking advantage of constantly evolving technologies for digital data capture, curation, analyses and dissemination. Digital data collection tools are perhaps the most rapidly changing arenas of development in cyber-archaeology and are becoming affordable tools for every archaeologist. In this paper, we examine two users' approaches to produce point cloud models of archaeological sites using structure from motion (SfM) photography. The experiment took place at the Fountain of Peirene in ancient Corinth, Greece. Their implementation of the technology and their results are compared to highlight the very important role the photo-shooting session can play in the final outcome of the SfM reconstruction. We correlate the users' approaches to the applied algorithms' robust features and known limitations to provide a technical explanation of how archaeologists can significantly improve their success in SfM. As new algorithms and software emerge making SfM a common tool in archaeological documentation the methodology presented in this paper will enable archaeologists to meet the high demand for digital documentation on a global scale.