Preface: Every so often I’ll walk around campus and talk to myself…resulting in a text-based glog post (like this) on what I’m thinking about in relation to anthropology, technology, Google Glass, and advanced documentation systems.
So here goes.
The difference between 3D and 3D.
Lately I’ve been wondering about a couple of things regarding technology and cultural heritage, chiefly: “How real is real enough when it comes to 3D data?” as well as “What type of 3D data are heritage practitioners using when they say they’re using 3D–e.g. LiDAR-derived data or photogrammetric?” Although they may result in the same “thing”, they are not the same “thing”, but the problem lies in what to call that “thing”? Lastly, how will all this impact 3D printing and the various “rights” that accompany it. We are truly pioneers in a brave new world.
LiDAR, the birth of a new resource, and trouble ahead…
When trained digital heritage documenters (like myself) gather xyzRGB measurements of a cultural resource with advanced technologies like a Terrestrial LiDAR Scanner (i.e. 3D Laser Scanner or TLS) the first end result is a 3D point cloud. This pointcloud, however, is far more than just simple measurements. When used in concert with one another, each point in the point cloud allows the whole to become “something else”. By digitally capturing reality–the ‘as-is’ state of something–we have created an entirely new type of heritage resource–Virtual or Digital Cultural Heritage. These pointclouds–typically when gathered by trained experts–are metrical accurate to anywhere from 2 millimeters to microns.
Because most of us in cultural heritage documentation are interested primarily in that as-is documentation, we are primarily concerned with getting data that’s again as accurate, precise, and representative as possible. The goal of every archaeologist is to systematically and scientifically learn about past human cultures, societies, and behaviors through the study of their material remains. We meticulously document every phase of our research in part to leave a record behind for other scientists to follow, under the auspices that our “documentation” is sufficient for them to recreate whatever it was that we were doing because–sadly–most archaeology is a destructive science. It was not until the advent of digital technologies that archaeologists were finally and truly able to recreate the conditions before, during, and after an archaeological excavation, for example. We no longer had to guess; to allow subjectivity into the record, we could objectively show our conclusions. Therefore, the pointclouds of cultural heritage objects or structures have the capabilities to completely represent–in detailed metric accuracy & precision–their parent resources. Typically after turning the pointclouds into those highly accurate and precise 3D models, one could completely recreate ANYTHING that was properly scanned.
Likewise, photogrammetry is a technique which utilizes photographs to create 3D models of whatever is depicted in them. Photogrammetry-derived 3D models (while increasing in metric quality) are nowhere near the accuracy or precision found in a LiDAR-derived 3D model. They, however, share the same name–3D model and this is where our troubles begin…
Finally, the third way to create a 3D model is to do just that–create it by hand from our human imaginations. You can see how subjectivity can quickly overtake objectivity.
Who owns the past?
With great power comes great responsibility, however, and the power held in 3D point clouds or the 3D models derived from said clouds is ridiculously immense. Websites like CGTrader.com allow users to take the 3D models they create (from whatever means–be they photogrammetry, scan-data, or by hand) and share them with others in addition to SELLING them. Likewise, anyone could create a 3D model of the Eiffel Tower and 3D print it at home…ponder this question:
What does it mean for museums, collectors, archaeologists, digital documenters, the American public, and for other global publics when artifacts and objects like the Venus de Milo, Lincoln’s Life Mask, or an Etruscan amphora could be photogrammetrically 3D-modeled, and 3D printed (or subsequently recast in bronze, or lathed from marble)?
So who owns the past–those people (both past and present) with cultural affinity and patrimony to an object or place–or those with the technological means to record and copy them? Conversely, who owns the modern (3D data)-past? Can a museum own the physical object and someone else own the 3D data? Do those who invent the means by which to document own what they create?
So, how “real” is real when it comes to 3D data?
This brings me back to the title of this post: how “real” is real; how “real” is 3D data? The current answer is it depends. Yeah, I went with the good ol’ archaeological catch-all answer because right now it’s true. There are no clear lines for what is real when it comes to this new digital cultural heritage because it is culturally relativistic. We cant treat it like a photograph when it comes to copyrights or fair-use because 3D scan data is far more powerful than that. It is “something else” and right now various institutions and governments are the ones in control of it, while various countries have different opinions about it.
The next issue is that not all “3D” is created equally. A photogrammetrically-derived 3D model will NOT (currently) have the same metric accuracy, precision, and representativeness as one created with a 3D scanner (and even scanner data varies depending on factors…) so when various entities say that they have “high-definition/highly-accurate 3D models” even that can be erroneous. If we are all working towards some sort of database for actually highly accurate/precise/representative 3D models then we must have a baseline understanding; we must have rules and peer-reviewed evaluation.
So how does all this relate to Google Glass…
Find out next week!