Here Comes the SCAPP Bot!

Editor’s Note: Here is a second post by/for Dr. Alexandra Ratzlaff. Alex has been working with the Brandeis Techne Group as Residents at the Autodesk Technology Center and in partnership with the Brandeis MakerLab run by Brandeis’ very own Ian Roy. They have some pretty amazing work to feature.

The overarching aim of the Brandeis Techne Group as Residents at the Autodesk Technology Center in Boston is to develop new equipment and methodologies to help push forward the collaboration between technology and the humanities. With a focus on archaeological research and applications, this group seeks to develop new ways of analyzing the material culture of the ancient and historical world. The initial goal of our project is to fabricate a prototype ‘Single Camera Automated Photogrammetry Platform’ (SCAPP) with the final designs and methodology available for reproduction through an open-source platform. The SCAPP is intended to be relatively low-cost and easily reproduced as an alternative to other digital imaging equipment.

Techne-Logo2

Archaeologists tend to be technological and equipment scavengers, often looking to related fields for ways to gather data and perform an array of methods of object analysis. The advantage of this is that it is a process and environment that breeds innovation. It is easy to work outside the box when you have very few limitations. This spirit of innovation is what drives our Techne Group.

71517370_2899107190119240_5067884211814268928_n (1)

After being initially approached by a member of the Autodesk team at the Boston Technology Center, Alexandra considered how to combine her interests in technology with archaeological field methods. With over 15 years of experience excavating, she recognized the importance of new technological applications used in digital imaging, mapping, and site virtual reality among other areas. The Autodesk Technology Center presented an opportunity to not only develop new equipment to be used in excavation and lab analysis but also a means towards further bridging the gap between the humanities and technology. A partnership with Autodesk and Brandeis also would have the potential to be used as a learning experience for students and a way to experience the humanities through a completely new lens.

IMG_2986

After spending months negotiating a contract between the two institutions, Alex then had to build a team. Having met Ian Roy the previous year, Alexandra considered his background in digital applications, imaging, and 3D workflows particularly relevant and ideal as a potential team partner. Ian’s extensive background in 3D scanning, photogrammetry, and photography, as well as participation in archaeological excavations in Greece as a technology director and specialist, all complimenting the potential collaboration with Alexandra and the Autodesk team. In an effort to make the Autodesk team a beneficial learning experience for students, Alexandra selected a graduate and undergraduate student from the Department of Classical Studies to join the team. Erin Brantmayer (MA 18’) and Helen Wong (BA 19’) contributed years of archaeological excavation experience as well as Erin’s previous work with field photogrammetry and Helen’s comprehensive work with 3D scanning provided the team with a variety of skill sets.

In an initial meeting with the early team members, numerous ideas were proposed in how to effectively integrate new forms of technology into archaeological methodology and what current applications could be further modified and improved upon? Ian showed the group the research of Professor Duncan Irschick’s “BEAST Cam”, a photogrammetry rig that employs approximately forty dSLR cameras to capture image datasets. This methodology produces high-quality images and models; however, the equipment is far too cumbersome for archaeological fieldwork and impractical for budgets as well. Ian’s suggestion inspired the group to focus on digital imaging. This resonated with archaeologists, Alexandra noted that a current problem in archaeology is obtaining high quality photogrammetry of artifacts as they are found in the ground before removal, in which some fragile artifacts sometimes further deteriorate. From this conversation, the idea emerged to develop and fabricate a new type of photogrammetry equipment that would incorporate the group’s interests and expertise.

However, it was evident that any development of equipment would require the input of a professional engineer. Ian proposed another member of the Brandeis RTI team, Tim Hebert, the Embedded Systems and Robotics Specialist and head of the Automation Lab in the Brandeis Library. Tim and Ian had worked together since 2013 as part of the team that founded the MakerLab at Brandeis. Tim contributes a background in embedded systems engineering and mechanical logic, as well as meticulous CAD design with Fusion 360. Together the initial group began their residency at Autodesk in the winter of 2018. Alex took the role of Team Lead and Principle Investigator, Ian took the role of Head of Method, Workflow & Planning, and Tim took the role of Head of Engineering and Design.

In archaeology we are constantly seeking new technological applications to field and survey research. However, very seldom is any of the equipment or software we use specifically designed for archaeology. Recording architecture and artifacts is a cornerstone in archaeological methodology, any developments in this area can become vital for the preservation and interpretation of a site or assemblage of artifacts. This was the broad problem identified by the group, further focusing on the development of equipment that could be used in the field and laboratory on objects that are highly specular or unsuitable for structured light scanning.

Copy of 2019-02-15 16.02.48

The SCAPP resolves these issues as a relatively light-weight and portable automated rig for a single camera. The design helps to rectify another issue faced by archaeologists in the field, which is how to perform photogrammetry on small object before they are removed in excavation. Some artifacts can be extremely fragile, such as glass or metal, the SCAPP is specifically designed to be put over an artifact to collect a photogrammetry dataset before the artifact is ultimately removed in excavation. If it is too fragile and further deteriorates while being removed at least a partial or full 3D model can be created from the SCAPP data. It also allows consistent and repeatable results. Our goal in the Techne Group is to “Science the past” – so much photogrammetry is based on artistic decisions, we want to “science” this problem. We have found that by normalizing our data acquisition, we can be agnostic about our processing methodology: cleaner data results in cleaner renders regardless of the software used.

 

The Techne group is truly a product of multi-disciplinary experiences and approaches to problem-solving. As our group keeps refining the capabilities of the SCAPP we will look at other issues facing archaeology in the realm of digital imaging and reconstruction.

Copy of 2019-04-05 14.41.25

The Process: What is a SCAPP bot?

Design and fabrication of an automated gantry system to move a dSLR camera in a full circuit around an object in order to take 30-100 + photos as datasets processed in photogrammetry software. The output generated through these programs provides a scale and color accurate 3D model, best for objects with high secularity.

  •  The initial prototype was constructed in birch to test the perimeters of camera angles and the feasibility of the geared frame and arch system.
  •     In the second incarnation of the SCAPP the entire frame was cut from aluminum using the Autodesk waterjet; carriages were printed in PLA and Markforged; and a computer from an industry standard “Ramps board” – the materials were chosen for their accessibility and relatively low cost.
  •     The SCAPP operates essentially as if it were a 3D printer, it uses a printer control board and motors, but moves and actuates a DSLR or phone).
  •     SCAPP can also function as a non-automated circular tripod in which data acquisition can be done manually while maintaining the leveled set degree position of the camera. 

SCAP Prototype

The Fieldwork

Designed as a portable imaging tool, SCAPP testing has included laboratory settings such as those at Autodesk, the classroom, and at an archaeological excavation. Initial SCAPP field testing was carried out in the summer of 2019 at Tel Kabri (israel) by Alexandra Ratzlaff and Erin Brantmayer of the Brandeis Techne Group.

While data was collected in the field, team members Ian Roy, Tim Hebert, and Daniel Lay developed an updated version of the SCAPP based on feedback from field and continued lab testing. Currently, the group is continuing to improve and refine the engineering and design of the SCAPP through collection of data sets primarily on artifacts from the Brandeis CLARC (Classical Artifact Research Collection). 

The Future

In archaeology we are constantly seeking new technological applications to field and survey research. However, very seldom is any of the equipment or software we use specifically designed for archaeology. Recording architecture and artifacts is a cornerstone in archaeological methodology, any developments in this area can become vital for the preservation and interpretation of a site or assemblage of artifacts. This was the broad problem identified by the group, further focusing on the development of equipment that could be used in the field and laboratory on objects that are highly specular or unsuitable for structured light scanning.

The SCAPP resolves these issues as a relatively light-weight and portable automated rig for a single camera. The design helps to rectify another issue faced by archaeologists in the field, which is how to perform photogrammetry on small object before they are removed in excavation. Some artifacts can be extremely fragile, such as glass or metal, the SCAPP is specifically designed to be put over an artifact to collect a photogrammetry dataset before the artifact is ultimately removed in excavation. If it is too fragile and further deteriorates while being removed at least a partial or full 3D model can be created from the SCAPP data. It also allows consistent and repeatable results. Our goal in the Techne Group is to “Science the past” – so much photogrammetry is based on artistic decisions, we want to “science” this problem. We have found that by normalizing our data acquisition, we can be agnostic about our processing methodology: cleaner data results in cleaner renders regardless of the software used.

The Techne group is truly a product of multi-disciplinary experiences and approaches to problem-solving. As our group keeps refining the capabilities of the SCAPP we will look at other issues facing archaeology in the realm of digital imaging and reconstruction.

A Digital Apolococyntosis

Introducing a new series (#SciencethePast): My colleague, Dr. Alexandra Ratzlaff, has been working with the Brandeis Techne Group as Residents at the Autodesk Technology Center and in partnership with the Brandeis MakerLab run by Brandeis’ very own Ian Roy. They have some pretty amazing work to feature, but in our autumnal mood, here’s a post-Halloween Update.

We posted earlier on the Pumpkinception, but here are some images and links to higher resolution models. We used the pumpkin exercise to practice some of the work we do with objects in the Brandeis CLARC (Classical Art Research Collection) and to train for work we do in the field (more on that soon).

We scanned the Techne pumpkin using an Artec Spider 3D scanner and then rendered it in Metashape.  (Here’s the Artec 3D website.)

Here’s a link to the Sketchfab version

For comparison, we did the same thing with photogrammatry using the SCAPP and rendered it in Metashape too.

What’s the SCAPP bot? Stay tuned…

If you go to this link, you can see the 3D model and some other cool stuff they are doing.