Modeling the Past: Photogrammetry and Anthropological Research
For my CAP project this year, I decided to do something at which I feel I’m particularly good: creating 3D models of artifacts found during CAP excavations. I have been using digital technologies to render 3D models for about three years now and have created these models for a number of reasons and for a number of different projects. In making the models, I use a technique called photogrammetry, which, at its most basic, takes a set of 2D images of an object, person, or place, stitches the images together, and if all goes well, a 3D rendition of that object is created in digital space. (A lot of other computer sorcery happens deep within the software’s code to actually create the models, of which I understand none, unfortunately.) The power of this methodology and its ultimate results cannot be overstated for a few reasons, of which I will discuss a few here.
1) Preservation and Conservation
As anthropologists, we crave to observe and document all that we see around us, or at least that which aligns with our research interests. However, try as we may, our data are not guaranteed to be preserved for eternity. We may accidentally lose our notes, delete photos, or objects might degrade or break over time. This last part is especially true when it comes to archaeologically recovered materials. We dig them up, clean them, study them, and try our darndest to keep them in good shape through proper curatorial practices, but the fact of the matter is that things break down over time. Photogrammetry is one way to counteract this phenomenon. Though not actually preserving the physical artifact itself, photogrammetry allows us as researchers to document materials in a way that we haven’t been able to before. Creating these 3D models gives us a digital copy of the object that can be manipulated and studied as if the object were there. Of course these models do have their limitations, such as not being able to look at it under a microscope and get to levels of specificity and detail that you may like. Nonetheless, photogrammetric models act as a backup that is more tangible that just photographs alone that may last longer than the physical object itself.
2) Analyses
Another important potential for photogrammetric models of artifacts is the ability to import the 3D files into different software packages and run various types of statistical analyses on them. The models don’t just look cool, they can function in an actual research capacity as well. Photogrammetry software (such as Agisoft PhotoScan which is what I use) allows you to scale your models within the program itself. This can be done by including scales within your photos that you use to create the model, then putting in the measurements of the scales into the program, thus scaling your model. By doing this, the research potential and usability of the model becomes significantly more powerful. One major additional advantage (which I will also touch on below) is that this means you can study the object without having it in front of you. Access to artifacts can be very tricky, but having a scaled 3D model of a particular artifact allows you to analyze it without physically needing it right there.
3) Ease and Potential for Data Sharing
One of the beautiful things about photogrammetry is that it is quite easy to do. Granted there are some rules you must follow in order to create an accurate model, but getting the hang of these is easy. In an ideal setting, you would have a lighting setup, proper cameras, and complete control over the space where you take your photographs. This is not always the case, but I have even had success in problematic environments such as in caves or places with no power. Photogrammetry allows you to take make models without having to lug around expensive and finicky equipment like 3D laser scanners, so the potential of gathering data becomes much easier. For example, myself and other students have gone on trips just to take pictures of artifacts to create models at different universities, museums, and even in different countries where the setup was less than ideal. Other modeling technologies would have been either cumbersome or non-functioning in some of these environments, but all you need for photogrammetry is a camera and the software.
Sharing data via photogrammetry is also incredible easy. You can export your 3D models in different formats and upload them to free and open source platforms, such as Sketchfab, where you can share your data and models with everyone. This becomes especially important for if you wish to show or teach about an artifact using a format other than a simple photograph.
The maneuverable 3D object is much more tangible and gives a more realistic sense of what that object is like. Some researchers have even uploaded different formats of models that allow you to make 3D prints of the object. This also has great potential for incorporating different technologies into the classroom, but also letting students (or yourself) physically touch a representation of an artifact that may be housed thousands of miles away.
I am trying to make at least one photogrammetric model of a CAP artifact every week so that we not only have a digital record of it, but also so the public can have a more tangible interaction with the artifacts themselves. We don’t have too many models up yet, but be sure to check out our Sketchfab account (linked here: https://sketchfab.com/capmsu) where we’ll be hosting all of our 3D models for the public! Also be sure to follow us on facebook, instagram, and twitter as we’ll sharing an ‘Artifact of the Week’ in 3D model form!