I participated in the2014 Mellon Institute in Digital and Public Humanities for Early Career Scholars, which featured theory and methods on digital archives using Drupal, markup of text using TEI, and 3D imaging using 123D Catch. As part of this program, I worked on a team that partnered with the Tennessee State Museum to photograph and build a 3D model of an artifact that plays a prominent role in the museum’s exhibit Slaves and Slaveholders of Wessyngton Plantation.

The Project

As part of a collaboration with the Tennessee State Museum, the 3D team created a 3D model of a Meerschaum Pipe that is part of the museum’s exhibit Slaves and Slaveholders of Wessyngton Plantation. The pipe was a gift from the Washington family to the overseer of their tobacco plantation. This exhibit faced the problem that the museum’s display case could only show a part of the pipe, which meant that at least half of the pipe’s intricate designs were hidden from view. Additionally, because the pipe is fragile, the museum wanted to create a way to display the pipe when the exhibit traveled to other cities in Tennessee. A 3D model would allow museum visitors to see all angles of the pipe’s delicate carvings without risk to the 200-year old artifact itself.

The Team

Robert DeHart, Curator, Tennessee State Museum
Bradley Daugherty, Graduate Department of Religion, @BJ_Daugherty
Danielle Picard, Department of History, @DRPicardHIS
Dan Fang, Department of English, @underlaugh
Cory Duclos, Assistant Professor of Spanish, Spring Hill College, @CoryDuclos
Kevin Vanzant, PhD, Mellon Postdoctoral Fellow

Outcome

Please check out the model here: http://www.123dapp.com/Catch/copy/2395142#

Click to see a mock-up of what a museum webpage might look like: http://drpicardhis.com/tsm3d/meerschaum-pipe/

The Process

After seeing the pipe in the context of the larger Slaves and Slaveholders of Wessyngton Plantation exhibit and meeting with the museum’s curator, the team took a series of photographs of the pipe to see what methods worked best for creating a 3D replica in the 1-2-3D Catch software program developed by Autodesk [link]

The software advised the team to shoot a loop of about 20 sequential photographs in small increments around your subject, and then shoot another loop from a different angle. In the background, the software suggests adding sticky notes around the object to give the software a guide to determine how the photos overlap.  The team attempted three different approaches in an effort to capture the image successfully.  The first two methods involved hiring a professional photographer.  For the third method, the team used an iphone and ipad to photograph the image.

Method 1 prioritized the placement of the photographer, not the object, and attempted to minimize background activity from all angles.  The photographer used a professional grade Canon 50D dSLR camera with a 50mm lens on a tripod, which necessitated a distance of almost ten feet from photographer to object in order to attain the best results.  The space disallowed a full 10 foot radius for the photograph to circle a stationary object, so the team tried to rotate the object itself, allowing the photographer the space needed to get the best photographs.  The team also tried to remove background interference by holding up grey cardboard behind the object throughout the entire series of photographs.  This method was unsuccessful, despite the high quality of the photographs.  The rotation of the object changed its orientation throughout the process, which prevented the software from identifying the necessary points of commonality to convert the individual pictures into a singular, three-dimensional image.

Method 2 attempted to capture the object by prioritizing the stable orientation of the object.  The object was never rotated on top of the table; instead, the entire table was relocated mid-way through the session to allow the photographer the necessary distance from all angles.  By moving the table, this changed the background for many of the pictures, so the team again utilized grey cardboard to create a stable background.  Method 2 was much more successful than Method 1 and resulted in a workable, 3-dimensional representation of the object.

In method 3, the team used an iPhone/ipad to attempt the capture and employed the 123dCatch app.  The method yielded surprising results.  The app could identify the object, alert the user to harmful vibrations, and request specific photos from all the necessary angles.  The team was not able to capture the pipe via this method, but did attempt various other objects (a hat, a lamp, and a piece of the 3-dimensional folk art (“Cowgirl Gum Drop Girl”)) and the results were encouraging.  The editing tools on the app are quite limited but the image itself was our best result.  The success of the app opens up the possibility of much more widespread use because of the ease by which the 3d capture can be attained.  From start to finish, the successful capture of an object through the app consistently took no more than thirty minutes.

Public and Academic Potential

• The online model created via method 2 (with the professional photographer) was edited in 1-2-3D Catch and can be embedded in the museum’s website or downloaded to a computer that can travel with the exhibit.
• In general, 3D modeling can also be used to produce a 3D print of the artifact. Such a model may be useful to an exhibit such as this one because museum patrons could physically manipulate the printed model without fear of damaging the original object.
• For researchers and educators, modeling can help bring more objects into the classroom or their research so that they and their students can answer a number of questions, including how an object may have been used, viewed at different angles, and preserved/damaged over time. Additionally, when used in combination with other technologies, such as 3D gaming which allows users to “walk” through a space, researches can try to situate objects in their original environments and contexts.