Skip to main content

Press Backgrounder: Art Historian & IBM Scientists Embark On Most Extensive Study Ever Done On A Single Work of Art

A Journey to Unravel The Mysteries Surrounding Michelangelo's (second) Pietá for More Than Four Centuries

July 1 - 01 Jul 1998: Florence, Italy -- In 1553, Asconio Condivi, author of "The Life of Michelangelo," wrote of the artist's second Pietá, "It is impossible to speak of its beauty and its sorrow, of the grieving and sad faces of them all, especially of the afflicted Mother. Let it suffice: I tell you it is a rare thing, and one of the most laborious works that he has yet done... He intends to give the Deposition from the Cross to some church, and to be buried at the foot of the altar where it is placed." Two years later, Michelangelo took a hammer to his own work, now known as the Florentine Pietá, and tried to destroy it. Stopped by his servant, the artist gave the unfinished statue away. It was later repaired and finished by Tiberio Calcagni, an otherwise undistinguished sculptor.

Today Michelangelo's Pietá, found in the Museum of the Opera del Duomo in the Cathedral of Florence, is the subject of a remarkable collaboration between a team of computer scientists at IBM's TJ Watson Research Center in Yorktown Heights, New York and eminent Renaissance art historian Jack Wasserman. By the end of 1998, Wasserman and the IBM researchers will have compiled nearly two billion bits of data to build a near-perfect 3-D digital replica of the Florentine Pietá.

The brainchild of Jack Wasserman, a distinguished historian of Italian art and Temple University professor emeritus, the study will shed new light on the mystery surrounding the Florentine Pietá, and lead to new technologies and applications for three-dimensional computer
representations of real life items. The final result of this study, says Wasserman, "will be a single visual virtual model of the Pietá that will be set up in a computer and enable me to carefully study this work of Michelangelo and draw some long-awaited conclusions.

The Mystery Behind The Florentine Pietá

The Florentine Pietá is a group of four larger-than-life figures carved from a single block of marble: the broken body of Christ is held up by Mary Magdelan, aided by Nicodemus above her, and the Virgin Mary to the right. Only the figure of Christ is finished, although the left arm has been broken and repaired and the left leg is missing. Mary Magdelan's face is only roughly blocked out. Nicodemus displays the unfinished features of the artist himself.

Michelangelo took up work on the stone when he was already in his 70s, without commissions and, as Condivi put it at the time, "for pleasure, as one who, full of ideas and powers, must produce something every day." It would be the second of three Pietás. The first,
now in the Vatican, is the only one of the three ever completed. Michelangelo was working on the third, now in Milan, when he died in February, 1564.

From depositions at the time, it was suggested that Michelangelo had intended the second Pietá to be his tomb monument. Why he chose to mutilate the statue is unknown, although several theories have been argued. After Michelangelo gave the statue up, Calcagni worked on it intermittently until his death a decade later.

The result, as Wasserman explains, is a statue by two artists, Michelangelo and Calcagni. This fired Wasserman's interest to learn how the statue was repaired, as well, he says, to answer some critical questions about what Michelangelo intended when he set out to destroy the work. According to Wasserman, "One of the many advantages of building this kind of computerized digital image is that one can turn it in a computer, inch by inch by inch, and study the statue from all possible vantage points."

Answering Century-Old Questions Through New Technology

Many 16th century discussions on the topic of ideal proportions in sculpture were geared toward understanding Michelangelo's style and perspective. For Wasserman, the proportion and details of the second Pietá are curious; some parts are excessively elongated, while others seem suspiciously small. "Michelangelo had a tendency to carve his work so the viewer could see it as he walked in arc-like trajectory from one side of the statue to other," says Wasserman. "The ability to stand each figure of this Pietá up straight without distorting the dimensions and
proportions would provide valuable insight into the question of what Michelangelo's proportions were like, his general concept of proportions, how he meant the work to be viewed at his tomb site, and, perhaps, his intent in taking a hammer to it."

In an attempt to answer these questions, Wasserman engaged a photographer to do a complete study of the work, as well as an Italian research institute to do a technological analysis. Neither analysis satisfied his expectations. In 1997, his search brought him to IBM because of its technological expertise which he felt could help him analyze the statue. Shortly thereafter, IBM, because of its technological innovation, agreed to invest its expertise in this project. Gabriel Taubin, head of the Visual Technologies Group at IBM Research and specialist in geometric computation and image-based algorithms and scientific data visualization, was approached to take on the project.

Not only was the task at hand of artistic interest, but it presented technological challenges to the IBM researchers that made it a fascinating project. These challenges, says Josh Mittleman, a member of the IBM research team, include "collecting the data efficiently, assembling it into a three-dimensional model that has the desired properties, and using it to render very accurate images under different lighting conditions, from different points of view. And for all these processes, the volume of data that we have to handle carries the problem beyond the scope of
existing techniques."

To begin, Wasserman drew up a list of objectives for the representation. These included wanting to be able to view the statue from different points of view in order to understand how it was meant to be seen in proper proportions; looking at individual parts of the work from different points of view to understand their construction; seeing unusual points of views, say from directly above; and, understanding where there were sections broken away and what the work looked like before the broken off pieces were reattached. Lastly, he wanted to see if he could reconstruct the placement of one limb, a leg, that was never replaced, using the existing marks on the statue.

Capturing And Organizing Nearly Two Billion Bits of Data

Digitizing the Florentine Pietá for a near perfect replica required taking nearly 1000 digital mesh photographs of this more than 2.5 meter-tall statue; the photos would have to establish a single and consistent distance from the camera to the surface of the sculpture so that they could be reassembled digitally into a accurate three-dimensional representation of the work on computer.

IBM, in conjunction with a company called Visual Interface Inc., used a Virtuoso shape camera (originally designed for plastic surgeons to take three-dimensional photographs). This six-lense camera takes six photos from slightly differing points of views by projecting stripes of light onto the statue, taking side by side black and white photos. From these multiple photos, a computer algorithm then computes distances to the surface and reconstructs the image in three dimensions. A grid of laser beams is also projected on the statue to facilitate alignment of the 600-plus photographs to achieve the three dimensional final figure. Texture and color information is captured by a color camera mounted on top of the Virtuoso.

To assure that the system would do the job, the IBM researchers ran a test run on a home made paper machier mock-up of the statue in the labs at Yorktown Heights. In early 1998, they spent 12 days in Florence working in the small unheated room of the Museum of the Opera del Duomo that housed the statue. They worked when the museum was closed, from six PM to midnight and on Sundays. "We worked at night and processed data during the day," says Rushmeier, "and our only chance to see Florence was when we went to the hardware stores to get odds and ends."

With no guarantees that the camera would be able to capture the entire surface of the sculpture, including the various nooks and crannies, between figures, IBM researchers basically took about 700 overlapping photographs of the statue. Not until returning to the IBM lab and
aligning all of the digital patches on computer was the team able to see whether there were holes. "In that first go-around, there were holes here and there, and those things were hard to get at," said researcher Holly Rushmeier. To fill the holes, Wasserman and the IBM researchers will
return to Florence next month to capture the final images.

While developing new hardware and software to manage the enormous amounts of data collected in digitizing the Florentine Pietá was challenging, IBM researchers see some of these methods being applied more generally to the problem of digitizing very large real-world
objects. This would enable one to examine an object in exquisite detail even without access to the original. In addition, new applications in the graphical representation of three-dimensional objects on the Internet, from displaying art work to allowing the manufacturing industry to produce on-line catalogs with three-dimensional displays of their products, may also be a possible outcome.

Short term, the IBM work on Michelangelo's Florentine Pietá will be compiled on a CD-ROM to be included in a book of essays on the sculpture that will be edited by Wasserman and published next year by Princeton University Press. "The main function of the book is to document the statue factually in as extensive a way as possible," says Wasserman, "and also to have analytical studies, theoretical studies--from historical, scientific and aesthetic points of view--of the various information gathered from the statue. The fundamental importance of the book is to make available a fully-documented presentation of the statue. The analysis are supposed to expose some of the questions and possible answers. These may be right or wrong but will at least open fresh areas of thought, with the idea that future scholars may add more documentation to it."

# # #

For more information on the the geometric modeling of the Florentine Pieta and visual art, visit, or for Italian.

Contact(s) information

Kendra Collins

Related XML feeds
Topics XML feeds
Chemistry, computer science, electrical engineering, materials and mathematical sciences, physics and services science