Some of the reasons to make 3D models of archaeological sites are

• preservation in case there's an accident (for example, the left front of the tomb's entryway has collapsed -- note the boulder on the left in the picture above that used to be part of the roof),
• documentation for remote access by scientists, and
• education by enabling students to examine the site without traveling there.

Methodology

We used the 3rdTech DeltaSphere rangefinder (pictured at right through the entrance of the tomb) to capture geometry.  The original prototype of the DeltaSphere was developed by Lars Nyland at UNC.  It is a time-of-flight laser scanner.  Up to 360 degrees horizontally and 145 degrees vertically can be measured at a resolution of up to 20 samples per degree.  We made most measurements at the full view possible, and at 13.3 samples per degree.  This results in over 10 million points per scan.  We took data from twenty positions, both inside and outside the tomb in order to try to measure as much of the environment as possible.

Color must be captured separately, and was problematic in such a dark environment.  We used a Kodak DCS330 camera mounted on the DeltaSphere in order to pan the camera automatically over the 360 degree view.  In order to light the scene, we used a Nikon flash.  This flash was able to light the scene satisfactorily, but resulted in some uneven lighting.  We are currently conducting research on automatic extraction of surface properties.

Results

The image below is a screen shot of the 3D model of the outside of the tomb.  It corresponds roughly to the view shown at the top of the page, although the virtual camera has been moved back to show more detail.

The image below shows a detailed view of the tomb entrance.

The following image shows a cutaway view of the inside of the tomb.  The entrance is to the left.  The near wall has been eliminated so we can see inside.

The tomb model has been adapted to the virtual environment system at UNC.  Visitors can wear a head-mounted display and walk around inside the virtual tomb.

Participants

Anselmo Lastra and Lars Nyland did the scanning.  Nic Terrenato set up the expedition.  Thanks to John Thomas for handling the shipping details and building some of the equipment, and to UNC graduate students Chad Hantak and Kok-Lim Low for helping with the equipment preparation.  Also to Professor Walter McCall of St. Mary's University.  He had a group of students surveying the walls of the town and provided a great deal of information about the site.  Finally, thanks to the members of the Archaeology Club of Rome who were on site digging and proved to be excellent companions.

To Lastra home page.

Last updated 05/31/06 .