4.4 Structure from Motion and Range Imaging

Figure 8: (VIDEO) Conceptual illustration of the documentation of an archaeological excavation; generating 3D textured meshes by Structure from Motion. Illustration based on data from the Alken Enge excavation (Holst et al. in press)

At present, an increasing array of techniques exists that allow easy, digital, spatial 3D representations of objects or entire scenarios. Some of the more prominent examples are the different photogrammetric techniques - especially Structure from Motion (SFM) (Wu 2011; 2013; Agisoft). SFM allows the creation of highly detailed 3D models based on photos alone, and is currently seen as a cheaper, faster and more flexible alternative to 3D scans, using dedicated 3D scanners (De Reu et al. 2013; 2014; Ducke et al. 2011; Koutsoudis et al. 2014; Pollefeys et al. 2001; Berggren et al. 2015; Powlesland 2014) . The technology is based on the discipline of range imaging and basically works by principle of parallax. Several photos of an object or a scene, taken from different angles, are automatically compared and matched two-and-two by similar feature-points to calculate the camera movement between individual photos and combined to a position in an arbitrary three-dimensional frame of reference. In turn the 3D positions of individual points are calculated, and potentially georeferenced from fix points in the photos (Figure 8). In its most basic form, the output of the method is a point cloud of 3D points, in essence comparable to the output of a laser scanner. However, the similarities end with the fact that the data output usually includes colour or even bitmap texture information, and that an entire scene is covered in one relatively quick photo session. The time it takes for a computer to calculate the 3D model poses a problem regarding a seamless integration with field recording, as it is likely to introduce a bottleneck in the workflow, caused by waiting for a result before the excavation can continue. It is, however, something that may be addressed through distributed and high-performance computing (Stott et al. in press). Different software, such as Agisoft Photoscan or Meshlab additionally allow for the generation of detailed meshes that are either coloured by vertex or texture mapped by face. Finally, perspective distortions (Escher effects) that would normally arise from ordinary photogrammetric rectification of not perfectly flat surfaces, are overcome using SFM to create orthophotos, which tie in nicely with usual archaeological 2D documentation.


Internet Archaeology is an open access journal based in the Department of Archaeology, University of York. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

Terms and Conditions | Legal Statements | Privacy Policy | Cookies Policy | Citing Internet Archaeology

Internet Archaeology content is preserved for the long term with the Archaeology Data Service. Help sustain and support open access publication by donating to our Open Access Archaeology Fund.