Digital Forma Urbis Romae Project

Fragment #010g (marble, 3 feet long, 150 pounds)

Click here for a 1669 x 1045 pixel (296KB)
or 3337 x 2090 pixel (775 KB) version of this fragment
(mosaiced from several high-resolution photographs)
A collage showing 1,163 of the 1,186 fragments that exist

Click here for a 1750 x 1188 pixel (598 KB)
or 3500 x 2376 pixel (2.2 MB) version,
or the 6800 x 4152 pixel original using Zoomify EZ

Versione Italiana

News flashes:
  • 6/1/05 - Our 3D models and photographs of all 1,186 fragments are now online, viewable by the public, and freely licensable to scholars.
  • 5/11/05 - Two new papers, one describing the entire project at a high level, and the other describing new fragment matches we have found.

    Overview of the project

    The Forma Urbis Romae, also known as the Severan Marble Plan, is a giant marble map of ancient Rome. Measuring 60 feet wide by 45 feet high and dating to the reign of Septimius Severus (circa 200 A.D.), it is probably the single most important document on ancient Roman topography. Unfortunately, the map lies in fragments - 1,186 of them, and these fragments cover only a fraction of the original map surface. Piecing this jigsaw puzzle together has been one of the great unsolved problems of classical archaeology.

    The fragments of the Forma Urbis present many clues to the would-be puzzle solver: the pattern of surface incisions, the 2D (and 3D) shapes of the border surfaces, the thickness and physical characteristics of the fragments, the direction of marble veining, matches to excavations in the modern city, and so on. Unfortunately, finding new fits among the fragments is difficult because they are large, heavy, and numerous. We believe that the best hope for piecing the map together lies in using computer shape matching algorithms to search for matches among the fractured side surfaces of the fragments. In order to test this idea, we need 3D geometric models of every fragment of the map. To obtain this data, during June of 1999 a team of faculty and students from Stanford University spent a month in Rome digitizing the shape and surface appearance of every known fragment of the map using laser scanners and digital color cameras. Our raw data consists of 8 billion polygons and 6 thousand color images, occupying 40 gigabytes.

    The goals of the Digital Forma Urbis Romae Project are threefold: to assemble our raw range and color data into a set of 3D (polygon mesh) models and high-resolution (mosaiced) photographs - one for each of the 1,186 fragments of the map, to develop new shape matching algorithms that are suitable for finding fits between 3D models whose surfaces are defined by polygon meshes, and to use these algorithms to try solving the puzzle of the Forma Urbis Romae. Whether or not we succeed in solving the puzzle, one of the tangible results of this project will be a web-accessible relational database giving descriptions and bibliographic information about each fragment and including links to our 3D models and photographs.

    This project is sponsored by the National Science Foundation under the name Solving the Puzzle of the Forma Urbis Romae. Some of the early work was funded under an NSF Digital Libraries Initiative pilot grant called Creating Digital Archives of 3D Artworks. Other early funding came from Stanford University, Interval Research Corporation, the Paul G. Allen Foundation for the Arts, the Mellon Foundation, the City of Rome, and Pierluigi Zappacosta.

    Current status of the project

    As of June 2005, we have assembled 3D models for all of the fragments, representing about 8 billion polygons, we have built a database giving scholars full access to these models, and we have created a separate database that gives the general public viewing access to the models. Click here for a description of (and access to) these two databases.

    Unfortunately, due to slight miscalibration of one of our laser scanners in Italy, we have high-resolution (0.25 mm) models for only about 800 of these fragments. For the remaining 400, we have a high-resolution model of the top (incised) surface and a low-resolution model (1-2 mm) of the full fragment. Although we no longer have funding to improve these models, we welcome proposals from any research group or institution that wishes to help us with this task. Since scanning large objects at high resolution will always yield datasets with slight calibration errors, developing principled methods for overcoming these errors would be worthwhile research, and good solutions would undoubtedly be publishable.

    In addition, we have found about 20-40 matches between fragments of the map (depending on how many you believe). These matches are described in this paper, to appear in Bullettino Della Commissione Archeologica Comunale di Roma in 2005. A description of the first match we found - in the Circus Maximus - is included in the second photographic essay below.


    Recent papers about the project:

    Fragments of the City: Stanford's Digital Forma Urbis Romae Project
    David Koller, Jennifer Trimble, Tina Najbjerg, Natasha Gelfand, Marc Levoy
    Proc. Third Williams Symposium on Classical Architecture, Journal of Roman Archaeology Suppl. 61, 2006.
    Computer-aided Reconstruction and New Matches in the Forma Urbis Romae
    David Koller and Marc Levoy
    Proc. Formae Urbis Romae - Nuove Scoperte, Bullettino Della Commissione Archeologica Comunale di Roma, 2005.
    Protected Interactive 3D Graphics Via Remote Rendering
    David Koller, Michael Turitzin, Marc Levoy, Marco Tarini, Giuseppe Croccia, Paolo Cignoni, Roberto Scopigno
    Proc. SIGGRAPH 2004.
    This is a paper about the ScanView system. We use this sytem in our public fragment database.
    A shortened version of this paper was the cover article in the June 2005 issue of CACM.

    Other available information

    Publicity about the project

    Photographic essays from the project

    Scanning the
    Forma Urbis Romae
    23 newly discovered fragments
    of the Forma Urbis Romae
    Carving and breaking the
    Forma Aedificii Gatesensis
    Analyzing the fragments
    of the FAG
    Includes pictures of
    the first match we found!

    On a more personal note, here is a photographic essay by project co-director Marc Levoy, narrating a week he spent on an archaeological dig in the Roman Forum, getting down and dirty with the topography depicted on FUR fragment #018a.

    Copyright © 1999-2004 Natasha Gelfand, David Koller, Marc Levoy
    Last update: