I have the opportunity to develop a proposal to work with Brown's visualization group - supercomputers, a CAVE for 3-D presentation, great computer imaging possibilities. I've come up with schemes involving presentation of artifacts and landscapes... but I'm sure there are other possibilities, either that someone's tried, or someone has considered trying. What's the state of the field, the most interesting work so far? What needs more research?
Visualization in digital humanities: what are the possibilities?(8 posts) (7 voices)
Steve, knowing your work and role at Brown, I suspect you're thinking specifically in the cultural heritage / museums realm -- but correct me if I'm wrong.
Richard Levy presented a great (and image-rich) paper at the Playing with Technology in History symposium, “3-D Interactive Worlds as Educational Tools for Understanding Arctic Life.” I'm reproducing the abstract below, and the paper's available in draft form on the site.
I liked it, personally, because it was about affective and subjective response -- not about the kinds of big-data or analytical verisimilitude projects we so often hear about with regard to supercomputing and viz. (Since I imagine you'll get a bunch of responses about those possibilities -- work like David Koller's -- I thought I'd toss this one in at the outset!)
Richard Levy: This paper explores how interactive 3-D worlds and computer modeling can be used to excite interest in traditional dwellings constructed by indigenous groups in the Canadian High Arctic. This paper explores the degree to which digital replicas of objects, houses and simulations can be used to support the study of Arctic history in interactive worlds. Also examined is the idea that virtual worlds can to evoke emotive, effectual knowledge in indigenous users. The discussion will be based on our experience with primary school and college students, and Padleirmiut Inuit Elders who experienced digital reconstructions of Inuit dwellings in a 3D virtual theater (CAVE) at the University of Calgary. Results suggest that virtual heritage environments may be useful in initiating and establishing discourses in archaeological interpretation, as well as assisting personal identity recovery.
I've also replied to your Twitter query, but I'll pass along the same information. The NEH Office of Digital Humanities has put together a Humanities High Performance Supercomputing Resource Page that you might find useful. The page may be found at http://is.gd/fVPSI.
There were three recipients of the special NEH/Department of Energy Humanities High Performance Computing program.
o The Perseus Digital Library at Tufts University for its project Large-Scale Learning and the Automatic Analysis of Historical Texts. The Perseus project will be using advanced computational linguistic technologies to experiment with the analysis of ancient texts for the study of classics and other fields. Additional information may be found at http://is.gd/fVSeM .
o The University of California, San Diego for its project Visualizing Patterns in Databases of Cultural Images and Video. UC San Diego will be creating computerized visualizations of large databases of digitized cultural heritage materials and performing statistical analyses on the data to discover new ways of studying art and culture. Additional information may be found at http://is.gd/fVS8Q .
o The Institute for Advanced Technology in the Humanities at the University of Virginia for its project High Performance Computing for Processing and Analysis of Digitized 3-D Models of Cultural Heritage. IATH will process previously-acquired raw datasets of culturally valuable objects such as artistic statuary, archaeological artifacts, and historical architecture in order to create highly accurate 3-D models for the study of art and architecture. Additional information at http://is.gd/fVQhC . David Koller, the PI, has also received a Digital Humanities Start Up Grant to continue work on his project. More information on this project at http://is.gd/fVS1m .
Steve -- I would definitely check out the Virtual Vault project. It is a collaboration between the Arizona State Museum and the Center for Desert Archaeology to create this amazing virtual version of their huge collection of southwestern pottery. They are using low-cost technology to create 3-D images of their collection. It enables both museum visitors and researchers to virtually "handle" the pottery that, in real life, is too fragile for you to touch. We helped fund it via a Start Up Grant. I saw Doug Gann from UArizona demonstrate it at the museum and it was really terrific. Doug is also experimenting with putting the virtual images in CAVEs and the like so you can compare them to each other over time. Here's an article with links:
There are a variety of things you can do with 3D visualization. I have worked on several cultural heritage reconstruction projects, most recently two Roman houses, the House of the Faun in Pompeii and the House of the Drinking Contest from the port city of Antioch. Both models recontextualize the artwork in the houses by placing high-resolution images of mosaics into the appropriate spaces and calibrating the model to the latitude and longitude of the site in order to accurately simulate the sunlight at a particular minute in history. This is especially useful for observing the way light interacts with artwork, architecture, and people, especially during social situations like formal dinners (example: http://people.virginia.edu/~ewg4x/hotdc/caa-hdc-medium.mov). The lighting simulation methodology can be applied to more than just these two projects, but to any site reconstruction. Candlelight can also be simulated. Applying accurate lighting to a reconstruction enables one to pose scholarly questions that otherwise never would have been considered. The next step is to bring the House of the Drinking Contest into a 3D game engine, Unity. Unity is becoming the standard engine for cultural heritage reconstructions. It is powerful and free for non-commercial purposes. Once the model is in Unity, the museums that now own the mosaics that were once in the house will be able to use the model for demonstration purposes to patrons.
I've also worked a bit with landscapes. Last year, I created a model of the Temple of Artemis at Ephesus, and the Austrian Archaeological Institute (who excavates Ephesus) graciously provided me with GIS terrain data for the square kilometer in which the temple resides. This terrain included a large mountain to the southwest. I converted this terrain data into a 3D model that I imported into my project, and with lighting simulation, was able to recreate the landscape more realistically, observing when the temple fell into the shadow of the mountain.
As for artifacts, there are a wide variety of 3D scanning projects out there. Depending on the type of artifact, you may not need to create 3D models to observe the three-dimensionality of the object. Polynomial texture mapping may be of use for objects that are somewhat flat, like coins or illuminated manuscripts (http://www.hpl.hp.com/research/ptm/)
I'm cooking up a project here at Carleton to use the web.alive platform from Avaya http://http://avayalive.com/WaStore/ (which uses the Unreal engine) to recreate the layers & contexts from a small scale training excavation, as a tool for demonstrating how archaeological knowledge is created. The idea being that be engaging with actual manipulable objects, deeper learning is created.
I'm also working with agent based modeling & simulation, using the netlogo environment http://ccl.northwestern.edu/netlogo/4.1.1/ . Models I've built live at http://www.graeworks.net/ . You could use Netlogo to prototype a simulation, and then scale up with the supercomputer... On the GIS and Agent Based Modeling blog, they also describe some experiments in marrying the two together (3d immersive worlds where you interact with artificially intelligent agents) http://www.gisagents.blogspot.com/.
Here is my provisional list of the ways visualization could be used in the digital humanities. Could DHers suggest more to add to this list, and/or better categories?
~ For language-based research: text mapping, word trees and clouds, and phrase nets (all for text analysis); scatterplots (for historical usage); databases (of texts and metadata).
~ For spatial/temporal research: timelines and historical maps (both with animations, embedded documents, and other related information);
~ For performance/fine arts research: simulations and animations of performances.
You must log in to post.