3Di or 3D Imaging from Free Range Photography

Hov3D_Sony_RX100

3Di Model of the eroded Anglo-Saxon Angel Frieze now protected within the church in Hovingham, North Yorkshire, (this version clipped on the right hand side). Generated using Agisoft Photoscan Pro from 20Mp images gathered using a Sony RX100 pocket camera

Over the past 4 years the LRC has undertaken a series of experiments designed to evaluate the potential for digital photogrammetery to enhance the  archaeological record from archaeological excavations. More recently a spell as Field Archaeologist in Residence in the McDonald Institute of Archaeological Research in the University of Cambridge enabled me to devote some time to a more in depth assessment of the potential of the method and to work with a number of others to showcase what I consider to be one of the most significant breakthroughs in computer applications for the field archaeologist.

The combination of a relatively high quality and high resolution digital camera with good lenses, a powerful desktop computer equipped with a gaming graphics card and software operating freely using cloud computing facilities or commercial software running on the local computer can generate very high resolution 3D Images or 3Di Models of excavated trenches and features with millimetre accuracy from a small number of carefully collected digital photographs. The 3D models resulting from this process enhance the record by revealing detail that cannot conventionally be recorded on paper, particularly relating to soil textures which can be enhanced by the use of carefully placed virtual lighting. In addition they can be manipulated on a web-page  or published within 3D PDF files in a format that supports measurement and model sharing in a way that is not otherwise possible. The product of the research in Cambridge will be more fully discussed in a Guide to 3D imaging for Archaeological Excavations due to be completed in June of this year.

In the meantime it has become very clear that the approach which employs sophisticated computing algorithms including Structure from Motion  offers tremendous potential in other fields in the Arts and Humanities beyond the relatively limited field of archaeological excavation recording. The potential to create highly accurate, shareable and measurable models for instance of Anglo-Saxon Carved Sculptural Stone or Inscriptions that can be shared between academics working in diverse locations around the world at the same time as making models that can be viewed online available to school-children or the retired has the potential to transform the  work of Historians, Linguists, Art Historians and others in the study and analysis of material much of which is inaccessible either because of its physical location in a building or simply because it is located in another country. Click on either of the two models pictured here to view the 3D models. These Models are large and thus may take a while to download and display if your connection is poor or computer is low powered. To be able to rotate the models correctly Google Chrome has proven to be the most consistent browser

Models generated of a recently discovered Runic Inscription were circulated amongst the community of runologists for comment over a few weeks rather than wait for years until each had had the opportunity to see the inscription in situ.

The experiments have been looking both at the process as well as the results this has included the comparison of both photographic equipment and software. To get an idea of the quality of the results you can view the experimental results on our Sketchfab site, (https://sketchfab.com/d.powlesland) which employs the Sketchfab viewer to display 3D models on a web-page or download some of the samples in PDF files, from the Cambridge University Digital Repository (D-Space) which provides a permanent digital archiving facility within the university. If you open a PDF from the D-Space it may open as a blank PDF in your browser with just the model title in the lower right corner; if this is the case download and open the file from your downloads directory using the Adobe PDF Viewer.

3D Model created using Agisoft Photoscan Pro from Digital Images gathered using a 16Mp Olympus OM-D EM-5

3D Model created using Agisoft Photoscan Pro from Digital Images gathered using a 16Mp Olympus OM-D EM-5

The 3D Models archived at Cambridge have all been generated using Agisoft Photoscan Professional (http://www.agisoft.ru/products/photoscan/professional) whilst those on the Sketchfab  include others generated using free online processing services available through the ARC 3D WebService (http://homes.esat.kuleuven.be/~visit3d/webservice/v2/) or Autodesk 123D Catch (http://www.123dapp.com/catch)

Advertisements
This entry was posted in 3Di 3D Imaging, Archaeology, art, Digital Heritage, Photogrammetery and tagged , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s