Procedural Modeling for Rapid-Prototyping of Multiple Building Phases
M.Saldana and C.Johanson
Presented Feb.25, 2013 at the 5th International Workshop on 3D Virtual Reconstruction and Visualization of Complex Architectures, Trento, Italy
Abstract (full paper online here):
RomeLab is a multidisciplinary working group at UCLA that uses the city of Rome as a laboratory for the exploration of research approaches and dissemination practices centered on the intersection of space and time in antiquity. In this paper we present a multiplatform workflow for the rapid-prototyping of historical cityscapes through the use of geographic information systems, procedural modeling, and interactive game development. Our workflow begins by aggregating archaeological data in a GIS database. Next, 3D building models are generated from the ArcMap shapefiles in Esri CityEngine using procedural modeling techniques. A GIS-based terrain model is also adjusted in CityEngine to fit the building elevations. Finally, the terrain and city models are combined in Unity, a game engine which we used to produce web-based interactive environments which are linked to the GIS data using keyhole markup language (KML). The goal of our workflow is to demonstrate that knowledge generated within a first-person virtual world experience can inform the evaluation of data derived from textual and archaeological sources, and vice versa.
At dawn on Sept.22, a few dozen people gathered to observe the Autumn equinox alignment at the Casa Rinconada kiva in Chaco Canyon. The sun rose behind a cliff and appeared though two perfectly east-west aligned doorways… the only time of the year this happens. No one knows if this alignment was intentional when the structure was built, but in 2012 at least, it generated participation in this natural phenomenon through the medium of architecture… as the park ranger observed, the event of the alignment is not a “representation” of the equinox or solstice — the man-made structure and act of observation is what causes the existence of “solstice” to come into being as a concept.
This summer I worked on procedural rules for a Hellenistic city. This particular city is Magnesia on the Meander, in Turkey, but the idea is that the models are parametric and their features and proportions can be easily adapted to reconstruct a number of other cities in Asia Minor. I wrote rules (scripts) for an Ionic Temple (the temple of Artemis Leukophrene, by the architect Hermogenes, is shown above); a Doric stoa with Ionic propylon; an enclosed altar (above); a theater/ stadium rule (this was a bigger challenge to do procedurally than it looks!) ; and simple rules for a bath-gymnasium complex and basilica. I tried to structure the code as a collection of ‘modular’ rules — for example, the Ionic entablature, colonnades, and pediment could be replaced by another order with just a couple of mouse clicks.
The scene is entirely geo-located and based on archaeological material collected in ArcGIS, as part of a NEH-funded digital survey being conducted by the Experiential Technologies Center at UCLA, and will eventually be viewable on Google Earth. The procedural models were created with Esri CityEngine. Next step: filling in the city grid with houses!
Theatron - from the same rule as the stadium, with a few parameters changed.
Temple of Zeus Sosipolis, in the Agora. This model derived from the same rule as the temple of Artemis, again with a few different parameters. Stoa with Propylon gate seen in the background.
View of the Agora with the Theatron and Stadium in the background. CityEngine is great for producing accurate, GIS-based topography.
The Humanities Virtual Worlds Consortium is a project sponsored by the Mellon Foundation that provides an online platform for gaming environments to be used in teaching and research.
Since 2011 I have been working with Prof. Chris Johanson of UCLA’s Classics department to create procedural content that recreates different versions of the Roman Forum in the republican period. We developed a workflow that starts with geo-referenced archaeological material in an ArcMap database. From this we can generate an accurate reconstructed topography, and locate the known buildings as well as devise a hypothetical layout of the republican city.
Using the procedural modeling software CityEngine, I wrote scripts that generate 3D models of the temples, basilicas and housing. The scripts provide a degree of randomness to reflect the unknown details, yet they are based on architectural ‘rules’ derived from a study of archaeological evidence of the period. The procedural modeling method allows us to create study models quickly and efficiently while controlling important architectural parameters.
We then used the 3D models to create multiplayer scenes with the game engine Unity. Through the Unity Web Player, the games can be streamed on the Humanities Virtual Worlds Consortium website, which also allows the creation of geolocated narratives and integration of textual arguments. This past spring, undergraduates in Prof. Johanson’s Spectacle Entertainments of Ancient Rome course created narratives within the virtual world for their research projects.
Big Church is a generative audiovisual project that uses photographs taken during a 2011 trip to Cappadocia. The program (written with Processing) randomly selects and distorts images of the landscape and interiors of cave churches, gradually disintegrating them using cumulative blend modes. This is a work in progress for investigating alternative ways of representing the dynamic of landscape and architecture in a mode that verges on the psychedelic. The title comes from the track by Sunn O))), from their album Monoliths and Dimensions, which provides the unofficial soundtrack for the piece and sonic inspiration (although in this version the visuals are not directly generated from the music)
Modeling Historical Cities in the Digital Age
(abstract from a paper presented at the Society of Architectural Historians 65th Annual Meeting in Detroit, April 2012)
Digital technologies have transformed the way in which we model and study historical cities. For archaeologists and architectural historians, models of ancient cities are hypothetical definitions of the invisible. Therefore, digital urban models must be malleable enough to change as new data comes to light, and the underlying theoretical precepts which are being tested must be explicitly stated and defined. This can be achieved through the interpretive modeling of variables derived from empirical urban theory. Empirical urban theory seeks to connect the actions of people to their built environment: it looks at how people shape the city, and how the city shapes them. It is characterized by systems of concepts that center on variables that can be quantified and tested. While it would be difficult to study the complex interrelations between these variables with a traditional urban model, new digital techniques such as procedural modeling now make it possible to provide useful definitions of the ways in which humans interact with the built environment. Procedural models are created according to parametric, conditional, or stochastic rules, wherein attributes and parameters are defined by the user to reflect empirical variables. The advantage of this approach is that models can be re-generated quickly and efficiently, allowing the second-generation model to be tested and evaluated, and so on until a defensible argument is reached. A case study that illustrates these principles reconstructs the architecture surrounding the Roman Forum in the mid-republican period.
Desert Sketches derives from a series of photos taken during High Desert Test Sites 2011 in Joshua Tree, CA. The images were altered with a program written in the Processing environment that obscured the details of the photos, extracting color and dissolving it into an abstract composition of lines and shapes in which some of the essence of the landscape is still perceptible. As a set, the images have lost the narrativity provided by people and contexts, while the blurred-out colors suggest a parallel, unconscious experience.
This project is an application made in the programming environment Processing that brings Bernard Tschumi’s Manhattan Transcripts into virtual, 3D space. The Transcripts are an exercise in alternative visualization techniques for architecture. Instead of being concerned with the mimetic depiction of space, they seek to investigate movement and experience through visual disjunction and the side-by-side juxtaposition of representational modes. In this spirit it seems natural to extend Tschumi’s experiment in the digital realm.
The application is not conceived as a “tool” for researching a body of work, but rather as a step towards extending Tschumi’s project into expanded dimensions. The exercise might begin to suggest other ways of transforming architectural experience and knowledge through media: for example, a research paper could be integrated into the application so that its text and argument are navigated visually and sequentially alongside the images.
Another aim of the project is to expose additional possibilities in Tschumi’s representational techniques by presenting the images in combinations and sequences not possible on the printed page or gallery wall. The sets of photographs, drawings, and movement diagrams can be viewed side-by-side or as semi-transparent layers thus making it easier to investigate the correspondences and differences between each representational mode. In addition, the images can be viewed all together or sorted by medium. Using the keyboard, the images can be cycled like frames in a film, highlighting Tschumi’s own references to the medium of cinema.