In order to explore the viability of a more rapid method of creating virtual reality re-creations of the existing remains (especially while excavations are on going) at Seyitömer and of still creating accurate and detailed textures for our models, the excavation team supplied us with many high-resolution photographs of the on-going fieldwork and architectural remains. Also, we took our own videos (at only 640x480) of two grain silos and then extracted 52 still frames from each video.
At various periods throughout the settlement's history, the inhabitants built numerous large outdoor grain-storage silos. Two of these were the subject of our photomodeling tests. The photos from the videos became the basis of the photomodeling stitching process. As our results showed (already back in 2009), this is a quick and inexpensive means of creating accurate and detailed interactive models (and in recent years has become more standard in many contexts, including at archaeological sites). Our models were then exported initially to VRML and (then in 2012) to 3D .pdf files for easy emailing, Web viewing, and real-time navigating.
As newer means of viewing interactives on the Web emerged, we continue to migrate the silo (and other) models. You can view them below using the Unity game engine exported out to WebGL.
The resulting models demonstrated that, in general, there is no need for expensive laser scanners, no need to deal with millions of vertices in a massive point cloud, and no need to decimate the model for ease of processing. Photomodeling provides a quick and inexpensive method of generating accurate and detailed interactive 3D models, just what a site like Seyitömer needed when it had to be excavated and recorded under severe deadlines.
By 2017, eight years after we introduced photomodeling to archaeological rescue operations, the technique has become commonplace. Many software kits now exist to expedite the process, create high-resolution interactive 3D models, and publish on the Web. Learning Sites continues to expand its use of photomodeling, as well (see e.g., our work at Jebel Barkal and our innovative automated cuneiform translator app).