3D Documentation and Visualization of the Forum Romanum:

Soprintendenza Speciale per i Beni Archeologici di Roma has been recently converted to the Soprintendenza Speciale per il Colosseo e l’Area Archeologica Centrale di Roma by bringing the central part of historic Rome under one umbrella under the direct supervision of the Ministero dei beni e delle attività culturali e del turismo (MiBACT).

The survey focused particularly on the most ancient of the fora, the Forum Romanum, from its westernmost edge at the Tabularium to the Temple of Vesta. Accordingly, the authors, when referencing the multiplicity of fora built during the Late-Republican and Imperial periods of ancient Rome, will speak of the ‘Roman Forum;’ when referencing the particular forum that is the focus of this study, the term ‘Forum Romanum,’ or simply the ‘Forum,’ will be used.

Giacomo Boni was the director of excavations in the Forum Romanum. Boni directed and documented this important project from 1898 until his death in 1925. Thus creating the most comprehensive documentation of the site till date. He did extensive studies in the stratigraphy of the Forum.

The original survey team in 2010 included (a) a digital survey group, composed of two graduate research assistants, Luke Golesh and Ryan Hughes, employing a Leica Geosystems ScanStation laser scanner; (b) traditional survey group, composed of four undergraduate assistants, Larissa Esmilla, Elizabeth Fuller, Kaitlyn Smous, and Lon Stousland, applying time-honored techniques of hand-measuring (c) a photogrammetry group, helmed by Prof. Dr. Selena Anders and supported by the undergraduate assistants; and (d) an ultra-high resolution photographic group, led by an undergraduate assistant, Ben Keller, using a GigaPan Systems EPIC Pro robotic camera mount under the guidance of Paul Turner.

Two further expeditions (2012, 2013) included teams like Prof. Giovanna Sandusky Lenzi, Dr. Christopher Sweet, Dr. James Sweet, and Ryan Hughes and undergraduate assistants Sharon McGolrick, Mason Roberts, Olga Bryazka, Keaton Bloom, Stephanie Escobar, and Taylor Stein. Offsite support for app and 3D design was provided by Markus Krusche, and Dr. Kristina Furse Davis. All surveys were conducted under the leadership of Prof. Dr. Krupali Krusche.

While the Forum history has been identified as started with the creation of Rome around 753BCE, new evidence suggests 900BCE existence of initial start of the Forum. Look at the study and excavation notes of Dr. Patrizia Fortini for the Lapis Niger.

Most notably, the Porticus of the Dei Consentes, the Shrine of Juturna, and the Temple of Vesta are contemporary reconstructions in which ancient fragments have been carefully combined with modern materials in order to evoke something of the monument’s original character.

After their discovery in the early twentieth century by Thomas Ashby, the drawings of Étienne du Pérac (also known as the Codex du Pérac) were very difficult to obtain. In 1960, however, Amilcare Pizzi published a volume containing these drawing with an introduction by Rudolf Wittkower.

Alongside an impressive body of built work, Andrea Palladio made a number of significant contributions to the field of architecture—most notably, the illustrations for Daniele Barbaro’s commentary on Vitruvius’ De Architectura and his own treatise, I Quattro Libri dell’Architettura are two publications by Andrea Palladio. The original illustrations of the first (1570) edition of the “Four Books” have been made readily available in a recent MIT publication.

While Desgodets advertised his work as having the most detailed and accurate drawings ever produced on the Forum, observations of the DHARMA team’s documentation reveal significant discrepancies between some of Desgodets’ renderings and the artifacts on site.

In his introduction to Ruins of Ancient Rome, Fillipo Coarelli writes:

“The result (of the Prix de Rome) is a body of drawings that, despite their unreliable reconstructions, often offer us invaluable records in diagrams, cross sections, and perspective views. This is not only because they provide irreplaceable images of Roman monuments made while the monuments were generally in better conditions but also because many times they are the only reliable sketches that were ever made. Their graphic quality is generally infinitely superior to the best produced in our own technological age….”

For example: Cairoli Fulvio Giuliani with Patrizia Verduchi’s surveys of the central Forum area; G. Foglia with G. Ioppolo’s elevations, plans and overhead plans of the Temple of Saturn and surrounding context; K.A. Nilson and C. Persson with Inge Nielsen and Birte Poulsen’s Temple of Castor and Pollux; Stefano de Angeli’s Temple of Vespasian; G. Pala’s Tabularium; G. Ioppolo with G. Foglia’s Tabularium; ADSAR with M. Cecchini’s Temple of Divus Julius; F.O. Shulze’s Temple of Divus Julius which includes five elevations; A.M. Ferroni’s Temple of Concord; H. Bauer’s Basilica Paulli; R.T. Scott with Groma’s Temple of Vesta; and F. Bianchi’s Temple of Antonius and Faustina.

A laser scanner is an automated surveying apparatus that uses Light Induced Detection and Ranging (LIDAR) technology to collect precise measurements of an object’s location. These coordinates are recorded in the form of a “point cloud” that reproduces an object’s shape by converting its spatial geometry into thousands, if not hundreds of thousands, of individual points, each with x,y, and z coordinates. This paper concentrates on the ‘time of flight’ type of laser scanner, the properties of which are specifically suited to historic and archeological documentation of large sites. The Cyra Technologies Cyrax 2400, designed by Ben Kacyra, became the first time of flight scanner with limited production in 1998. It consisted of a single high-speed laser beam that has a return signal at the same angle of coincidence capturing data via oscillating lightweight mirrors. Leica Geosystems acquired Cyra in 2001 and has since highly improved the scanner’s performance and rate of data capture.

Although not discussed at any length in the present paper, these traditional techniques were critical to the DHARMA team’s work on the Forum. While ultimately playing a role supplemental to that of the digital survey, these techniques provided irreplaceable experiences of the Forum site. A comparative evaluation of digital and traditional techniques may be found in the Conclusions section below.

Gigapixel Panorama, or GigaPan, image is a super-large, digital, panoramic photograph, which is composed of a series of smaller digital images. This method of data capture allows GigaPan to capture expansive vistas and simultaneously achieve astonishingly high levels of detail. GigaPan technology, initiated by Carnegie Mellon University and NASA Ames Intelligent Systems Division’s Robotics Group, functions by using three basic systems: GigaPan Imager System, Stitcher software, and the GigaPan website. The GigaPan Imager is a pivoting robotic camera mount used to install the digital camera and to auto click images based on a selected window frame, set zoom, and with a specified number of vertical and horizontal images. The GigaPan Stitcher software is a custom-designed to combine these large sets of digital images with the least amount of parallax and best exposure levels in order to create the final ultra-high resolution panoramic image. The GigaPan website is the platform, by providing both the unique storage capacities required to handle these data-rich images and the viewing features (most significantly, pan and zoom) necessary to take advantage of their richness, needed to host such data files.

Comprehensive, technical information, including comparative analysis- both lab and field tested- is available from the symposia from the International Scientific Committee for Documentation of Cultural Heritage (CIPA), and in the Journal of the International Society of Photogrammetry and Remote Sensing (ISPRS).

A datasheet of the Leica Geosystems ScanStation model laser scanner is available for download at CyArk website at http://​archive.​cyark.​org/​leica-scanstation-data-sheet-blog.

This is according the Leica Scanner datasheet (see CyArk) and uses a Guassian laser.

An example of this method may is available from the team’s work of the Forum. With the scanner located at ground level, the top of the entablature of the Temple of Antonius and Faustina was a much greater distance from the scanner, than its base. We thus divided a scan that reflected a1.5 cm × 1.5 cm resolution of a distance of 28 m (the distance to the top or the entablature) down a third of the column shafts. The second scan was of the middle third of the column shafts. A 1.5 cm × 1.5 cm resolution was again used but at a distance of 25.5 m (the farthest distance within this scan). A third scan of the base of the shafts, a 1.5 cm × 1.5 cm resolution was acquired at 17 m. Finally, ground scans were taken and 13 m and 6.5 m.

This static method of data collection could, indeed, be considered a deficiency of current scanner technology. A future growth of the technology may include “dynamic” resolution, in which the scanner is capable of acquiring a uniform resolution for the model regardless of object distance.

While it is possible, albeit challenging, to register individual scanworlds without targets during post-processing, the team insisted upon the accuracy and convenience natural to target-based registration.

More precisely, the site was traversed, capturing the major monuments, in the following order: the Porticus of the Dei Consentes (scan position 1), the façade of the Tabularium; the Temple of Saturn; the Temple of Vespasian; the Temple of Concord; the Arch of Septimus Severus; the Basilica Iulia; the open, central space (the east and west rostra, the small monuments around the latter, and the Diocletianic columns); the Curia; the Basilica Aemilia; the Temple of Divus Iulius; the Temple of Antoninus and Faustina; the Regia; the Temple of Vesta; and the Temple of Castor and Pollux (scan pos 27).

A 1.5 cm × 1.5 cm “adaptive” resolution was implemented in order to produce a final model with accuracy up to 1/2 in.

Accuracy, as proven by field tests at the University of Notre Dame, was no problem at this extended range.

As seen in Fig. 5, the pivots occurred at scan positions 8, 14, 19, 24, 25, and 26.

The massive base of Temple of Castor and Pollux and relatively flat topography surrounding it forbade simple capture. In addition to those already planned, two pivot scans and ten target locations were required to acquire the necessary data.

A more detail report on the methods employed to create integrated methods of study are seen under, Krusche, K. reference articles.

While importing the model into these software programs is possible, the large size of the model currently requires hefty processing power.

More recent scanning devices, with each new generation of Leica scanners, have greatly increased the speed of data capture with associated benefits and drawbacks.