We begin scanning the David

Photographs and text by Marc Levoy
February 17, 1999

On Monday, February 15, we began scanning Michelangelo's giant statue of David. At 8:00 in the morning, a crew arrives to remove the glass barricade from the statue. This job is harder than it sounds; each pane of glass is an inch thick and weighs several hundred pounds. An hour later, a second crew arrives to construct a scaffolding, which will serve many purposes for us.
It's early afternoon, and the scaffolding is being used by technicians from Florence's restoration laboratory to give the statue a thorough dusting. David has been waxed several times during his history, and the restorers would like to know which parts of the statue are still covered with wax. Although it's hard to tell a waxed surface from a polished surface, wax fluoresces when illuminated with ultraviolet light, so in addition to scanning David's geometry and color, we plan to scan him under UV light, make a map of the waxed areas, and project it onto our 3D model. Unfortunately, dust also fluoresces under UV light, so the statue must be dusted before we begin. Click here to see a result from our ultraviolet scan.
Meanwhile, in another part of the gallery, our crew has been laying out the pieces of our gantry on the floor and bolting them together. With ropes and strong arms, we raise it into a standing position, then load the base with 400 pounds of counterweights to keep it stable. Using pulleys, we hoist the horizontal arm and the scanner head to the top of the scaffolding and bolt them to the upright gantry. Here's what the gantry looks like after the scaffolding has been pulled back. From the ground to the top of the scanner head is 24 feet. Note the broad platform at the bottom of the gantry. It sits partially on the gallery floor and partially on the stone pedestal of the statue. This platform was constructed specifically for scanning the David.
It's evening now, and with the setting of the sun, the skylight above the David darkens. It's time to begin scanning. We position the scanner head about 3 feet away from the statue and start sweeping it with our laser. In a few minutes, the laser will turn off, a calibrated white light will turn on, and the same area will be scanned for color. In about 30 minutes, this portion of the chest will be finished, and we'll reposition the gantry for another scan.
The David looks different close up than it does from the ground. This picture was taken while standing on the scaffolding. From this distance we can see the thin marble ridge Michelangelo used to emphasize the boundary between David's lips and his face. The famous furrowed brows, which look natural from the ground, actually protrude from the forehead in a way that is anatomically impossible. Note also how the gaze directions of his two eyes diverge. For more details about this unusual feature of the statue, click here. While we're admiring Michelangelo's genius, the laser is busily working its way through the hair on the left edge of the photo.
We'll be scanning David from 7:00pm until 8:00am every night for the next few weeks. There are 22 people on our scanning crew. Everyone (even the prof) works roughly one 7-hour shift in the museum each day. People who aren't scanning are back at the laboratory processing data, writing software, or sleeping. It's a Herculean labor. Here's the crew, from left to right: Kathryn Chinn, Jeremy Ginsberg, Sean Anderson, Lucas Pereira, Dan Perkel, Matt Ginzton, Alex Roetter, Wallace Huang, Maisie Tsui, Marc Levoy, David Koller, Unnur Gretarsdottir, Kari Pulli, Alana Chan, Dana Katter, Daniel Wood, Rahul Gupta, Szymon Rusinkiewicz, Brian Curless, Jelena Jovanovic, and James Davis. Missing: Semira Rahemtulla.
Oh yes, I almost forgot. In order to mount the scanner head on the gantry, it must mate closely to the scaffolding, yet not collide with it. To explore this problem, an accurate scale model (15:1) of the gantry (on the left in each photo) and the scaffolding (on the right) with its pulley system was built by Marc Levoy and Benjamin Levoy (age 7) from K'nex parts.

To read more about our marathon scan of the David, including a compilation of statistics about the dataset, click here.


© 1999 Stanford Computer Graphics Laboratory
levoy@cs.stanford.edu