Photographs and text by Marc Levoy
April 20, 1999
|
|
On Monday, March 29, we move from the Galleria dell'Accademia - our home for the last two months, to the Medici Chapel, where we'll stay until early May. The day begins at 7:30am with the arrival of two truckloads of scanners and computers at the front door of the church of San Lorenzo. We have to unload them fast, because at 8:30 Florence's famous outdoor market opens, making vehicular access impossible. Note the church's rough facade. Michelangelo was commissioned by Pope Leo X to design a marble facade for San Lorenzo, but it was never completed. The horizontal courses of protruding stones were intended to support this facade. |
|
|
With the help of our team of professional artwork movers, the scanners and computers are wheeled down the nave of the church, past the altar, and through a side door leading to Michelangelo's chapel. The move goes surprisingly smoothly, despite the necessity of removing a massive Renaissance-era wooden door from its hinges to admit the gantry base. There are few things you can't move with enough strong arms. You can see the door propped against the wall of the church. |
|
|
By late morning, we have pretty much filled up the chapel with our equipment, which is closed to the public today. In this shot, we see the tomb of Giuliano di Medici, topped by the reclining allegorical statues of Night (on the left) and Day (on the right) and in the niche above them a statue of the duke himself. Behind us is the matching tomb of Lorenzo di Medici and the allegorical statues of Dawn and Dusk. Although everything in the chapel was designed by Michelangelo, including the walls, it is hotly debated how much was actually carved by him. However, few historians doubt the authenticity of the four allegories, and scanning them is our primary goal. What is that wooden structure at the right edge of the photograph? Well... |
|
|
Early on Saturday morning, January 2, a 20-pound chunk of marble detached from the interior of the chapel dome and fell 100 feet to the floor below. Fortunately, the chapel was closed, so nobody was hurt. Also fortunately, Michelangelo's statues are along the walls, so none were damaged. To repair the damage and restore the dome, this giant scaffolding tower was built in the center of the chapel. It rises vertically for 60 feet, then spreads out to support a wooden floor (visible at the top of this photo) that closes off the dome from the rest of the chapel. |
|
|
At first we worried that this tower would impede our work. Certainly, it cramps the narrow space of the chapel floor. However, climbing up into the tower, we soon realized that it provides us with a historically unique opportunity - views of the chapel walls that nobody has ever seen. With the permission of the superintendency, we are hoisting our Cyra scanner up into the tower. By scanning the walls from vantagepoints like this one halfway up, we hope to build a computer model of unprecedented detail and completeness. By the way, climbing this tower is not for the faint of heart. |
|
|
Climbing to the top of the tower, one enters an ethereal space between the temporary wooden floor and the interior surface of the dome. Does the pattern of concentric square coffers remind you of the Pantheon in Rome? Most art historians think that Michelangelo modeled his dome after this famous ancient example. From here, another scaffolding continues upwards... |
|
|
...until one is standing inside the lantern itself. The name is misleading; there are no light bulbs here. The Renaissance "lanterna" was a disk cut from the center of a dome, raised on columns to form a cylinder, and enclosed with glass. Its purpose was to bring as much light as possible into the dome. Here is a view looking straight up into the lantern. |
|
|
The gash in this column on the side of the lantern shows the location of the 9-inch piece of marble that fell to the floor below. The expansion of a rusting iron pin is thought to have weakened the marble, but the triggering event is suspected to be the loud bombs that traditionally accompany New Year's eve fireworks in Italy. The piece fell the day after the fireworks. |
|
|
Here is an outside view of the lantern and the dome. The lantern, designed by Michelangelo, was widely praised by his contemporaries for its simplicity and slender elegance. |
|
|
Descending back into the chapel, we see our Cyberware laser scanner working on the statue of Dawn. These statues are smaller and lower than the David, so we have removed several pieces from our gantry tower. It's a good thing, because space is tight in the chapel. These statues are also more highly polished than the statues in the Accademia. This has forced us to increase the intensity of our laser, search for scanner positions that avoid specular reflections, and filter our data afterwards to reduce the artifacts that are caused by subsurface scattering. |
|
|
Here's a view taken from behind our workstations. On the left is a Silicon Graphics (SGI) O2, which we use to control our scanner (visible in the background). On the right is a more powerful SGI Octane, which we use to align scans and entertain tourists. Between them are some of the 18 gigabyte disk drives that we hand-carry daily back and forth between the museum and our laboratory across the river. |
|
|
While our Cyberware scanner is busily digitizing statues, our Cyra scanner peers through the bars of the tower at the chapel walls. Although the tower gives us unique views of these walls, it does not give us all views. Building a complete model from these scans will be challenging. Converting this model from a sea of polygons to a structured model that is useful in CAD applications will be even harder. |
|
|
A few feet away from the Cyra scanner, Stanford sophmore Maisie Tsui sits at our "aerial workstation" controlling the scanner, aligning scans, and chatting by Medici-Chapel-net to her teammates 50 feet below. |
Although it's nice to have a change of scenery after spending two months in the bustling, brightly lit Accademia, the Medici Chapel is a melancholy place to work. The walls are ringed with false windows and empty niches, the surfaces crawl with fantastical garlands and grotesque masks, and the tortured figures of the four allegories stare into space from blind (pupil-less) eyes. The middle-aged Michelangelo, himself obsessed with death, was successful in making this place feel like a tomb. Moreover, with the dome closed off and most of the spotlights dark, the only light filters down from windows placed high on the walls, creating the illusion of being deep underground. There's a chill here that our portable electric heaters don't dissipate.
On Wednesday, May 6, we rotated the monitor of our SGI Octane workstation to face the tourists, cranked up a program that let them navigate our partially completed model of Dawn, and stood back to watch. Here's what happened.
|
|
|
|
|
|
|
|
| They | came, | they | saw, | they | played. |
Kids of all ages. The program we gave them was an experimental multiresolution viewer written by Szymon Rusinkiewicz. The user was allowed to rotate the statue, translate it, zoom in or out, and move the virtual lighting, all using the mouse. Motions were constrained to prevent people from "getting lost", there were no on-screen menus, and the keyboard was not available. Instructions were posted (in multiple languages) nearby. The screen was located about 10 feet away from the real statue of Dawn.
Some observations:
On the surface it seems ludicrous to place a computer in front of a statue, and on the computer screen to display a 3D model of that same statue. In reality the computer seems to refocus the visitor's attention on the statue, and it allows them to view it in a new way. In the case of Michelangelo's statues, which are large, the available views are always from the ground looking up. Michelangelo knew this, and he designed his statues accordingly. Nevertheless, it is interesting and instructive to look at his statues from other viewpoints. In general, by allowing museum visitors to explore a statue themselves, or change its virtual lighting, we made the viewing of art an active rather than a passive experience. For at least a few hours, the art museum became a hands-on museum.
Note added May 2009: Inspired by the success of this experiment, we were invited by the Galleria dell'Accademia to install an interactive kiosk next to Michelangelo's David. The kiosk became operational in October, 2002. So far as we know, it's still there. Here is a web page about that kiosk.