Virtual Sambor Prei Kuk, Cambodia

The convergence of the desire to re-tell the history of SPK and the advent of web-enabled immersive, interactive new media technologies have provided us with the opportunity to develop a new way of experiencing cultural heritage. It is based on digital modeling of the site, the people, and the activities that were part of this heritage.

The guiding metaphor for such modeling is the concept of Place . A “place” is a setting that affords the entire spectrum of human activities, including physical, social, and cultural activities, while affecting, and being affected by, those activities. We chose this metaphor to guide our work because it pertains to both physical and non-physical settings. On one hand, place is often used to describe the territory that we occupy. The boundaries of this territory are defined by a sense of being inside—inside a region, a town, a neighborhood, a building. But that boundary is identified not by a demarcation of its edge, but by the feeling of coherence of the spaces, objects, and activities within it, which give rise to a competence in the way a place is inhabited. We value such places because they give us a feeling of being somewhere as opposed to just anywhere .[1] The physical attributes of the place frame the activities that ‘take place’ there, and provide its inhabitants with a socially shareable setting for their activities and rituals, in terms of cues that organize and direct social behavior that is appropriate for that particular place.[2]

To help us organize the model, we chose to compare our virtual place-making to a stage-play, which comprises a stage (a context), a narrative (the play), and actors (which include the audience, or visitors, in our case). The notion of place as a stage-play drives literary works, films, video games, and architecture. It provides a framework for understanding the individual contributions of the components, and their mutual interactions. Figure 1 illustrates the components and their relationships.

Figure 1. Diagram of the Stage Play metaphor.

The ‘stage,’ or context, comprises both space and time. It affords spatial and temporal grounding for the entire model, and includes spatial components like buildings, trees, topography, sky, etc., and ‘props’—objects that can be manipulated by the actors or can act on their own (animals, fire pits, the Linga, etc.). The ‘actors’ include avatars, or PCs (player characters), which are human characters controlled by the people who are logged into the system (the visitors); as well as agents, or NPCs (non-player characters), which are pre-animated, semi-autonomous entities, that perform pre-scripted roles, and sometimes have action modification capabilities based on some sensory input (e.g., they can start some action sequence when an avatar approaches them). The ‘play,’ or narrative, includes both cultural heritage aspects, and the activities that take place in the environment (known together as simulation/action). They tell the story (or stories), and afford the freedom to participate in the story.

The interactions between these components are what make them a ‘place’: the avatars, which are the representations of the visitors, can ‘see’ other avatars (as well as the other components of the game), and be ‘seen’ by them. Likewise, the NPCs can be seen by the visitors, and can react to their presence. This reaction both conveys some of the essence of the cultural heritage (they can perform actions related to the history of the place), and add to the authenticity and ‘sense of place’ of the experience. And of course the context (buildings, marketplace, etc.) help locate the experience, both spatially and temporally.

 
1. The ‘Stage’

The first problem facing digital reconstruction of a cultural heritage site is finding the appropriate documentation that describes the built environment and the ‘props’ for the period being reconstructed. In many cases, some buildings exist. They can be photographed, measured, or digitally scanned, providing a basis for the reconstruction. But much has been obliterated by time and nature.

Our modeling of the site was facilitated by a laser scan, provided by Dr. Takeshi Oishi of the University of Tokyo, in Japan (figure 2). This technology uses a radar-like device, which shoots a laser ray at the object being scanned, and records the time it has taken the ray to reach its target. The result is a point, defined in 3-dimensional space. Since the device repeats this procedure thousands of time every second, the result is a ‘cloud’ comprising millions of points, which together define the surface of the object. This point cloud is then processed through software to generate a three-dimensional model of the object.

Figure 2. Laser scan and model of SPK central shrine.

The site model was developed in 3DStudioMax (a modeling software made by Autodesk corp.), and exported to Torque (a game engine made by Garage Games Corp.), which powers our virtual world. Like other similar engines, Torque incorporates a physics engine, whereby ‘gravity’ is imposed, solidity of objects can be enforced, and time of day and weather phenomena can be included. Torque also provides mechanisms to support PCs and NPCs (player characters and non-player characters), which were useful for implementing the actors, as discussed in the next section.

 
2. The ‘Actors’

The second main challenge in reconstructing SPK was modeling the people who inhabited the site. The physical challenge has been mostly technical: modeling human beings is difficult, because we are so accustomed to seeing them in real life that any discrepancy is immediately, and disturbingly, obvious. Yet, to render high fidelity models of people would slow down the interaction, due to hardware limitations. We needed to develop a wide range of characters, both avatars for the player characters and ‘bots’ for the non-player characters (the NPCs), who would resemble some of the real people who inhabited SPK in the 7^th century AD. We relied on carvings in various temples, including Angkor Wat, and on literary sources that described the people of the region. And we chose to sacrifice some quality in order to gain speed.

 
3. The ‘Play’

Finally, the component that brings everything together, is the activity, or narrative. What does the visitor do in the virtual world environment, and how are all those actions and interactions tied together in the larger experience of the virtual world and the story it tells? In the case of SPK, we tried to do this by creating both small, localized activities, centered on various rituals (such as the market, the fire sacrifices, the ceremonial pool, and the linga ceremony in the central shrine), and the larger, overall narrative, which takes the visitor from outside the temple (the market) all the way into central shrine.

To maintain the ‘sense of place,’ visitors must choose an avatar (a representation of themselves) that fits the place. Hence, they appear in the scene in the form of one of the people who inhabited the site. Their activity is limited to observing the ongoing activities: they cannot partake in them. But they can chat with one another, via the chat mechanism provided in the game.

More pages:

• The Model
• Screenshots