Construction of Dressing and People
One of the greatest struggles for the reconstruction team was to figure out a way to give a sense of the daily life in ancient Jerusalem within the constraints of a real-time simulation model. Though desirable, a realistic computer model of a single human figure could easily surpass the number of polygons found in the entire Herodian Temple Mount model. (The Digital Michelangelo Project at Stanford, for example, has a dataset from a laser scan of the David statue that contains 2 billion surface polygons. In contrast, the entire Temple Mount model has less than 1.2 million.) Needless to say, it was impossible to fill the Herodian streets with walking and talking figures and still maintain any degree of interactivity.
The reconstruction team decided instead to position two-dimensional figures throughout the model - the computer equivalent of putting paper dolls in a three-dimensional dollhouse. The problem then became locating historically acceptable images that would also work visually within the model. Initial tests were done with figures from the frescoes of the third century synagogue in Dura Europas, the oldest known depictions of ancient Jewish life. Unfortunately, the results were visually jarring, so this approach was abandoned. Images of costumed actors were ultimately located and digitized. (The 'people' in the foreground of the model were photographed while filming the high definition video prepared for the Center; the 'people' in the background are courtesy Archives & Collections, Universal Studios.)
To create the texture maps, each figure was isolated in Photoshop and saved with a transparent background, or alpha channel. In MultiGen, this texture was mapped onto a single polygon that rotates on its central vertical axis. These types of files are known as billboards and will always rotate to face the user during the simulation experience.
Approximately 115 individual people were created for the Herodian Temple Mount model. Along Tyropoeon Street, the figures were placed individually in scenes designed to suggest the daily life of the city. To create the crowd on top of the platform, a file containing 35 figures was externally referenced multiple times and randomly rotated to avoid an obvious repeating pattern. In addition to the people, chickens, goats, mules, market wares, and awnings were also created and placed in the model to enhance the reconstruction and provide scale for the monumental Temple complex.