Blog

A Giant Ground Sloth is 2 ½ Fourth Graders Tall

By Bernard K. Means, Director, Virtual Curation Laboratory at Virginia Commonwealth University

Go to any public archaeology event or even an academically oriented archaeology conference, and you will probably see at least one button, shirt, or even a ball cap emblazoned with an image of a dinosaur (probably a Tyrannosaurus rex) contained within a thick red circle and bisected by a thick red line.

Below or surrounding that image might be the words “Archaeologists Don’t Dig Dinosaurs.” I use this image myself on my first day of lecturing in my Introduction to Archaeology course each semester. One of the basic messages that I try to impart on that first day is that archaeologists are not paleontologists.

3D scanning a bone from an enslaved context at VMNH pictured is Dr Elizabeth Moore
3D scanning a bone from an enslaved context at VMNH pictured is Dr Elizabeth Moore. Image by author.

We might like dinosaurs or other prehistoric animals, but if those animals lived before or did not interact with humans or their hominid ancestors, those “terrible lizards” and other ancient creatures fall outside of our purview. I also warn any budding archaeologist that they will get asked, probably by a relative and certainly by a member of the public, if they have “found any dinosaurs yet?” after their first, second, third, ad nauseam, dig.

However, I’ve begun to rethink over the last few months that image of the dinosaur struck out in red.

This reassessment of the phrase “Archaeologists Don’t Dig Dinosaurs” has grown out of my work in the Virtual Curation Laboratory with various natural history museums, notably the Virginia Museum of Natural History (VMNH) in Martinsville, Virginia, the Western Science Center (WSC) in Hemet, California, and the Las Vegas Natural History Museum (LVNHM), not surprisingly in Las Vegas, New Mexico.

Beginning at VMNH, I found fossils creeping into what I 3D scanned. Paleontologists at that museum asked if I would not mind scanning the occasional fossil in between artifact scans, such as Miocene whale vertebrae, a lungfish bone from Chile, or various Ice Age animals. In fact, for VMNH right now, I am 3D printing mirrored 3D scans of giant ground sloth bones—VMNH has lefts and rights and needs rights and lefts.

3d scanning a prehistoric camel skull at WSC
3D scanning a prehistoric camel skull. Image by author.

I’ve actually been 3D scanning and 3D printing quite a few Ice Age animal bones over the last year. In addition to the aforementioned giant ground sloth bones from VMNH, I have 3D scanned giant beaver (VMNH), mastodon (VMNH, WSC, and a private collection), mammoth (VMNH, WSC, LVMNH), short- faced bear (WSC), American camel (WSC, LVMNH) and dire wolf (LVMNH)—yes dire wolves were once real and not just something made up for Game of Thrones. These animals are all now extinct—and these fossils do potentially represent an overlap between archaeology and paleontology as these animals all lived at the time after humans entered the Americas. Some scenarios even argue that humans are partly or solely responsible for the extinction of these Ice Age megafauna (e.g. big animals!). But, the dinosaur bones I 3D scanned at LVMNH or the dinosaur tracks I 3D scanned at VMNH well predate the presence of humans in the Americas or anywhere across the globe, as dinosaurs went extinct millions of years before humans first evolved.

3D scanning an allosauroid vertebra at LVMNH
3D scanning an allosauroid vertebra at LVMNH. Image by author.

 

So, why bring up dinosaurs, or even the fossils of Ice Age animals from contexts not associated with human activity? This is because I increasingly find myself in settings that are less archaeology oriented, but more focused on how 3D technologies—particularly 3D printing—can help inform the past. And, I am talking to audiences that are generally interested in the past—whether the human past or an even deeper past. I want to encourage and celebrate this interest, this fascination with science, and not be quite so pedantic about, ”Well, no, I don’t do dinosaurs, ask someone else.” Paleontologists, as well as archaeologists, are also interested in public outreach and share our goal of teaching people about the past—our disciplinary differences are important, but I think we need to sometimes blur these boundaries.

3D scanning maeve foster
Scanning Maeve Foster.

Which gets me to the title of this blog. On May 1, 2017, I was invited to speak to the soon-to-be retired Judith Fitzpatrick’s fourth grade class about applications of 3D printing, and I brought along a wide selection of recent 3D prints, including artifacts from Pennsylvania, Virginia, and India, animal bones from archaeological and modern contexts, and various fossils, including a giant ground sloth claw. I also had with me a 3D print of a 4th grader—Maeve Foster—who had interviewed me the previous month for her class, as she is interested in 3D printing. Maeve was out ill that day, but her fellow classmates were fascinated that she was there in miniature 3D-printed form, and they placed her figure on her desk. The students really liked the opportunity to handle the replicated objects, and to discuss their significance.

3D printed giant ground sloth claw
3D printed giant sloth claw. Image by author.

 

maeve foster 3D
3D printed Maeve Foster. Image by author.

They were also glad to find out that they could download and 3D print some of the objects from our Sketchfab site. One of the items that I passed around that was of great interest was a 3D printed giant ground sloth claw, 3D scanned originally at VMNH. A young 4th grader asked how tall giant ground sloths were, relative to 4th graders. 4th graders not being a standard measure in archaeology or paleontology, I asked if one of the students would offer up their height. A young girl quickly stated that she was 4 feet tall. So, this makes the extinct giant ground sloth 2 ½ 4th graders tall.

setting up 3D prints at Short Pump elementary school
Setting up 3D prints at Short Pump Elementary School.

I certainly think the future of public archaeology—and public paleontology—as well as educational applications will be found in the chirping and beeping of the 3D printer. For more on educational applications of 3D printing, I refer the reader to this recent blog: Enhancing archaeological research with 3D printing.


In the Virtual Curation Laboratory, a team of Virginia Commonwealth University undergraduate students and alumni works under project director Dr. Bernard K. Means to digitally preserve the past and share it with the world. Check out and download digital artifact models on our Sketchfab page.

Printing a Plantation: Using Photogrammetry and 3D Printing to Bring Archaeology to the Public

Nestled among the dense forests and sprawling agricultural fields of Gloucester, Virginia is a little-known early Colonial property called Fairfield Plantation. Patented in 1648, Fairfield was the home of the Burwell family, one of the wealthiest and most politically influential families in Colonial Virginia. Their property reflected their prominent role in society, featuring a striking manor house and numerous outbuildings positioned at the heart of a substantial plantation. The house grew and evolved with the family over time, and remained home to the Burwells until they sold it in 1787. The property exchanged hands several times in the years that followed, until 1897 when the house burned and was left to ruin. Over a century later, Fairfield Foundation archaeologists are bringing new life to this important historic property using the latest innovations in 3D technology.

Figure 1
Historic photograph of the Fairfield Plantation manor house. (Courtesy of the Colonial Williamsburg Foundation)
Figure 2
Archaeologists working with Adventures in Preservation workshop participants at Fairfield Plantation. (Courtesy of the Fairfield Foundation)

The Fairfield Foundation is a non-profit organization headquartered in Gloucester, Virginia that has been promoting and involving the public in hands-on archaeology, preservation, and education activities within Virginia’s Middle Peninsula and surrounding areas since 2000. Their primary research site is Fairfield Plantation, where thousands of students, interns, and volunteers have learned about archaeology and colonial history at Fairfield Plantation, making it a valuable educational resource within our community.

Earlier this summer we initiated a project utilizing 3D technology to digitally record, reimagine, and recreate the historic landscape at Fairfield Plantation. Our goal is to develop an interactive 3D printed model that will bring the experience of archaeology to the community, and ultimately draw more attention and visitation to the site. Using a Phantom 4 Pro drone (affectionately named Major Tom), we have begun documenting the ruins and surrounding landscape by flying over the site and capturing hundreds of photographs, which are later transformed into highly detailed 3D models using Agisoft PhotoScan. These models will later be 3D printed to develop a tangible replica of the site.

Figure 3
3D model of a collapsed chimney at Fairfield Plantation. (Courtesy of the Fairfield Foundation)

What makes this project unique is that instead of having one solid model, we will be printing each test unit individually and repeating the documentation and printing process over time so that each layer we excavate in the field can be incorporated into the printed model as a removable piece. Members of the public will be able to take the model apart layer by layer and experience the same process of discovery that archaeologists do. We will also use the digital model as a basis for digitally reconstructing the house, which will be printed and incorporated into the replica. This replica will bring Fairfield Plantation to life, providing residents and visitors to Gloucester a chance to interact with the past and connect with local history. When finished, the model will be housed and publicly accessible at the Fairfield Foundation’s headquarters, the Center for Archaeology, Preservation, and Education (CAPE) in Gloucester Courthouse.

Figure 4
Ashley McCuistion flies the drone over the manor house ruins at Fairfield Plantation. (Courtesy of the Fairfield Foundation)

This project challenges people to experience history in a new, tangible way, and brings Fairfield Plantation into a local and global spotlight. Digital models and printed replicas of the site will be an integral part of lesson plans we will make available to individuals and classrooms around the world, drawing new attention to the rich history of Virginia as seen through Fairfield Plantation. It also brings the Fairfield Foundation new opportunities for public outreach and education, and places the organization at the forefront of a growing digital preservation movement in archaeology.

For future updates about this project, visit our blog at www.fairfieldfoundation.org/blog, and check out the 3D models we’ve produced of Fairfield Plantation at www.sketchfab.com/fairfieldfoundation.

 


Ashley McCuistion is an archaeologist with the Fairfield Foundation, a non-profit organization based in Gloucester, Virginia. In addition to being in the field, she is the Public Outreach Coordinator for the organization and Project Manager for the Fairfield Modeling Project. Ashley received her B.S. in Anthropology from Virginia Commonwealth University and her M.A. in Archaeology (pending thesis completion) from Indiana University of Pennsylvania. She has been researching creative ways to incorporate 3D scanning and printing technology into public archaeology since 2012.

Nephelococcygia (Cloud Watching) for Public Outreach

One of the limitations I often face when using 3D technology for public outreach is the still very small database of models to pull from. The process for creating a 3D model of an artifact or site is time consuming and most of us simply do not have time to model everything we would like too. Therefore, it is in our best interest to find ways of making 3D technology more appealing to our peers in other branches of archaeology. For this post, I tested the CloudCompare software and found that it has a lot of potential.

Testing the Program

Recently, the American Civil War monument that I used as a subject for my first photogrammetry model was knocked over. The Sons of the Confederacy group here in Tallahassee, Florida believe that it was accidently hit by a work vehicle and was not intentional vandalism. I saw this as a perfect opportunity to test out CloudCompare and see what changes might have occurred to the monument.

The CloudCompare software overlays the point cloud you get from either photogrammetry or laser scanning the original model

on another version

to get a very nice visual comparison of the two. This is the first version using CloudCompare:

Note the blue in the center of the monument, indicating little change, and the red edges, suggesting a lot of change. For this first comparison, I aligned the base and this revealed that the marker’s position had changed (it is now set more square than before). This is something that I had missed while photographing and looking at the models separately. If you click on annotation 3 we can very nicely see the new footprint for the marker on its base. This also might explain why I could not get CloudCompare’s auto-alignment feature to work like I was expecting.

Also in this comparison you can see the two large chips (marked by annotation 1) that were removed from the base. While these show up, comparing the base is problematic as the original model had a lot of grass growing right next to the stone, and the recent version has the glass cleared away. This resulted in a lot of false positives when looking for damage on the base.

Since it was the marker that was knocked over though, I really wanted to see how that had changed. After aligning the face of the models, I got some more interesting results.

The first thing you might notice is that the top edge of the plaque and the edge of the oval on the plaque are red. I suspect that what we are seeing is one version of the models not generating this edge very well.

Annotation 2 highlights another piece of damage that I had previously missed in the form of a new notch in the marker’s edge. If you look back at the new version of the completed model, these notches look pretty uniform and this marker has been moved before. This makes me wonder if the equipment used to hoist the stone is causing this damage.

Also note that the cosmetic damage to the plaque does not show up. I was hoping that the program might track color changes as well, but you cannot have everything. Also, a chip missing out of the top-front edge that is not highlighted as much as it probably should be. I suspect some fiddling with settings should make this pop out more.

Conclusion

As a test, this monument was the perfect subject. I saw the changes that I expected to see, saw changes I had missed, and learned about some of the limitations of CloudCompare. The potential for this program as a research tool for documenting sites and artifacts is pretty obvious. However, I see a more direct use in our public outreach programs as well.

In particular, the Florida Public Archaeology Network has a program called Heritage Monitoring Scouts where we are organizing public volunteers to help monitor coastal sites that are threatened by sea level rise. If we were to train certain volunteers to take pictures for photogrammetry purposes, we could use CloudCompare to help document the degradation of these resources over time. This would be very useful information and I suspect that participants would find it rewarding as well.

Any other ideas? How would you use this software?


Tristan Harrenstein is trained as an archaeologist and has a passion for outreach and education. This is fortunate as he is also an employee of the Florida Public Archaeology Network, an organization dedicated to promoting the preservation and appreciation of our archaeological resources. He shares a blog space with his boss (Barbara Clark) and you can read more blog posts on other archaeology related subjects that tickle our fancy here.

Incorporating 3D Models into FPAN Curriculum

At the Florida Public Archaeology Network’s East Central office we have been hard at work adapting 3D modeled and printed objects into curriculum based on artifacts recovered from the Kingsley Plantation National Park Site, a former slave plantation in northeast Florida. FPAN and the Kingsley Plantation host an annual workshop for teachers to introduce Project Archaeology curriculum and the site-specific supplemental, Investigating a Tabby Slave Cabin . The supplemental curriculum uses the archaeological investigations of one of the plantation’s slave cabins as the framework for discussing how archaeologists investigate shelter.  One of the educational aids used in teaching the curriculum is a floor mat based on plan drawings of the site. Teachers and students then use paper cutouts of artifacts recovered from the site. While this is a neat idea, the problem is that paper cutouts are not terribly engaging to adults, let alone children.

01.JPG
Participants in the Project Archaeology workshop that takes place at the Kingsley Plantation work together to recreate an archaeological site to learn how archaeologists investigate the concept of shelter. Here, they are using cut-outs of artifact drawings to replace them into their original context.

A real improvement to the curriculum could be made if the paper cutouts were replaced with 3D printed models. FPAN requested access to a portion of the Kingsley Plantation artifacts from the National Park Service which were housed at the University of Florida. This portion of the collection is currently a part of ongoing research of the site under the direction of Dr. James Davidson, who directed the archaeological research upon which the FPAN curriculum is based.  One thousand photos in eighteen sets later, I was ready to start on the real work of photogrammetry

(Left) author Jeffery Robinson collects pictures for processing. (Right) Example of 360 round-table set-up for photography of artifacts.

The program that I used for photogrammetric rendering was Agisoft Photoscan, as it is the standard program for this use in the archaeological community. Now, because we had used a lightbox to shoot the artifacts in, masking the photos became a necessity. Trying to process without the masks caused Agisoft to render the model incorrectly. I also found several other reasons for errors in rendering: sharp edges, flat surfaces, reflective surfaces, and differences in lighting. Due to this and other errors, only twelve of the eighteen photosets were able to be completed. Those models can be viewed, and pending NPS review, downloaded  here.

04 (1)
All of the completed models as shown on sketchfab.com/fpan. The models will be available for view/download pending NPS review.

After uploading, I printed out the models on a Makerbot 5th Generation Replicator. This is a great machine for printing out good quality models and is pretty user friendly too. After a short experimentation period to get it set up right, I was able to print out all of the models. The models were then sanded down with a dremel to provide a nice even surface. The models were then painted to provide more realism, as no one wants to look at a boring grey model.  Hopefully we will soon be able to send kits of these objects out to go with the curriculum sometime in the future.

(Top left) 3D models are printed on a Makerbot 3D printer. (Top right) the printed, plastic 3D models. (Bottom) The author applies acrylic paint to the models.

 

Jeffery Robinson has a B.A. in Anthropology from the University of Central Florida. He has been volunteering at FPAN’s East Central office for the past year.

Autodesk ReMake – Review

Over the last couple years, we have seen the 3D archaeology community grow at an incredible rate. As awareness for the potential of these tools has grown, I have been asked several times about the best way to get started. Depending on their need, I generally recommended Agisoft PhotoScan as the current standard in archaeology circles. With PhotoScan they can get lots of help, flexibility in subject scale, few limitations, high accuracy, it is relatively easy to use, and it has recently come down in price.

However, to anyone currently looking to get into photogrammetry for interpretive and educational purposes, I have a new recommendation. Autodesk ReMake is perfect for any program that just needs to make a model of an object without investing too much time and money.

Pricing

The professional version of ReMake currently runs at $30 a month or $300 a year. You could compare this to PhotoScan’s regular professional license of $3499, but they are really not related. A more appropriate comparison is to PhotoScan’s standard license which is currently available for one-time fee of $179. As a result, ReMake really does not make financial sense at this level.

However, ReMake has a free or educational license which makes it worth considering. There are features that are available in the full version that are not in these licenses. For example, neither can freely use “Ultra quality” images (though I have not had any issues) and the free version is limited to 50 photos (though you can do an awful lot with 50 photos). For the rest of this post, I will be talking about these two licenses.

Making a Model

If you want to get an idea of what it takes to make a model in PhotoScan, I suggest checking out the guide I assembled here. ReMake simplifies this process immensely. You merely select your photos, name the model, decide if you want their auto crop or smart texture options, and then hit “Start.” As we are working with the free or educational licenses, the program will then upload your photos to a cloud, process them, and then notify you when your model is available to download.

It really is that simple. There are some limited editing options for a finished model, but I have not had much need for them. As you can see, the results are excellent and more than serviceable. Occasionally, I have gotten ReMake to assemble a model that PhotoScan could not figure out.

 

Features/Limitations

For ReMake features and limitations go hand-in-hand. If you are using one of the free licenses then you only have the option of using the cloud for processing your model, you cannot create it on your computer (locally). If you are like me, this is perfect because my computer struggles to do anything while a PhotoScan model is processing. Also, unlike programs such as 123D Catch you still retain full rights to your model.

If you want a fully 3D object in ReMake (one with a completed bottom) then you will have to use a second program to assemble it like Blender or MeshLab. Technically, you can do this in PhotoScan which automatically stitches two models together to create a complete model. In practice, this has been far more finicky so I do not expect to miss it.

Finally, there is no masking option in ReMake. This means that, if your background is not out of focus, you will run into problems. On the other hand, this will cause difficulties in PhotoScan too, though you can overcome them with effort. That being said, the model below turned out pretty well, and this is one that PhotoScan struggles to make sense of.

 

ReMake is a much simpler and more accessible program than PhotoScan. This naturally means that you have less control over your model and you have fewer options. If a model does not turn out well, for example, you do not have the option of spending hours trying to manipulate the program to get a result (which actually appeals to me somewhat). On the other hand, if you have a casual, non scientific project where you just need a model for a demonstration or for printing, it is excellent.


Tristan Harrenstein is trained as an archaeologist and has a passion for outreach and education. This is fortunate as he is also an employee of the Florida Public Archaeology Network, an organization dedicated to promoting the preservation and appreciation of our archaeological resources. He shares a blog space with his boss (Barbara Clark) and you can read more blog posts on other archaeology related subjects that tickle our fancy here.