A cool new weird spicy kind-of-augmented reality application comes from a Berlin artist called Julius von Bismarck: he hacks your reality when taking a picture. For you unnoticed at first, it will reveal your screwed up sweet spot touristic shot only once you take a look at your frame! Check out his website and see the video and description for his device there.
Cam-Trax has released a video for their camspace project. They do real-time 2D tracking of arbitrary objects using off the shelf webcams. The quality looks pretty stable and even the fastest movements (throwing a rod through air, flipping it) won’t kill the tracking. Take a peek look at the game demos (AR Pong, etc.) and wait like me for the beta release to try it out. Otherwise it won’t be possible to judge about quality, i.e. what happens with changing/moving background? Lets hope that their pinch gesture to simulate mouse clicks works without colored patterns on their fingertips later on as well.
All documentation to the Augmented Reality Music Production project (by Daniel Hienert) are now online as well! Play augmented drums on a stereo projection screen and use AR technology to compose music, play instruments, mix up your tracks and replay midi files to learn by watching and doing… Check out the media section and the videos in Project: AR Musicproduction and get the paper under Publications. Enjoy!
Finally, I put up my documentation and the videos to the “Augmented Television” project. The idea is to overlay television images with additional annotations and 3D graphics in real-time - but not a priori or at the broadcaster side, but instead at the very end of the chain: I do the overlay inside the television set of each viewer in their homes. This opens up a whole new bunch of opportunities. Customized television for each user and full interactivity are now possible. Read/view on in the Projects AR-TV section and the Publications part. The work was shown first in May 2006 at the EuroITV in Athens.
Oxford University presented their AR tracking at ISMAR 2007. Their approach for marker-free environment feature tracking looks pretty promising. Be sure to check out their website and the latest videos.
