JUST IN: A documentary about this project, broadcast on the Big Ten network.
The Champaign Urbana Community Fab Lab contributed to a special Augmented Reality installation for the University of Illinois graduation ceremonies.
See the official press release and ‘about‘ page, and the students’ pix.
The Alma Mater statue is being restored, which created a crisis: this would be the only year for decades in which graduates could not get their picture snapped under Alma’s wide arms.
The campus powers that be wanted to do something, and asked Alan Craig if it would be possible to us Augmented Reality to put Alma–restored–on her pedestal, so people could get a snapshot. Alan and I conferred, and the conclusion was YES WE CAN, and more over, WE MUST.
The image was based on high resolution scans of the statue in the restoration process (which the University should have done for documentation anyway), cleaned up and rendered by the Beckman Institute Visualization Lab led by our colleague Travis Ross.
Besides my own contribution as software integrator, Andrew Knight of the CUCFL created a custom rig to hold an tablet computer on a camera tripod. You can buy similar products, but why would you buy one when you have a Fab Lab? 🙂
I was particularly happy to contribute to this project because it is a great example of using contemporary IT to create experiences that strengthen our social and human bonds. This installation was meaningful mainly as part of a social ritual (largely invented bottom-up by generations of students), and could only be experienced–together–at the specific time and place.
And I learned a bit about the particular challenges of life size (larger than life size in this case), outdoor, Augmented Reality.
I was surprised how well the AR worked. The large target far from the camera worked as well or better than small versions in the lab. The outdoor lighting didn’t seem to be a big problem, though obviously there are lot’s of conditions that will fail.
On the other hand, it was difficult to test, since the site was not under our control, and wrangling the rig was not trivial. So we had only one actual test of the final product, with no chance to iterate. Fortunately, it worked well enough we didn’t need to iterate.
We also found a number of technical problems we need to understand, such as how to map the virtual geometry to a complex setting in the real world. More on these issues later.
I should note that Joel Steinfeldt of the University Public Affairs Office was the key organizer who made it all happen. This project was a collaboration of many departments at the UI, and had a really tight schedule (with a very immovable deadline), so many things had to come together just right to get enough working to be useful. Joel was essential to this process: well done!
Robert McGrath's Blog 