No images in this article but interesting enough to post…seems one of the first advertising displays which counts how many people react to it…basically an early stage of our Intelligent-Display.
Well if you dont know the guys at nui-group lemme just bring your attention to them.
they have some great projects in there lineup, and are definetly worth a visit:
This years UX week in san francisco showed off lots of interesting stuff.
Check out http://www.uxweek.com/
an especially interesting presentation which focuses on gestures and touch interfaces can be found here:
Hardly anyone knows that a secret tunnel runs deep beneath the Atlantic Ocean. In May 2008, more than a century after it was begun, the tunnel has finally been completed. An extraordinary optical device called a Telectroscope has been installed at both ends which miraculously allows people to see right through the Earth from London
Simple…but awesome! Love it!
It seems like consumer-grade 3D Displays have finally arrived. Two of these babies started shipping in Japan these days. But on your 3D Shades and start watching. Not sure about the quality and realism those things provide…but finally a step into the right direction….I certainly will keep on waiting for my holodeck 
Sources:
http://www.engadget.com/tag/stereoscopic
http://www.engadget.com/2008/04/10/worlds-first-46-inch-stereoscopic-3d-tv-from-hyundai-on-sale-in/
http://www.dlp.com/hdtv/3-d_dlp_hdtv.aspx
The french site http://www.extreme-studio2cine.fr/ incorporates a text-to-speech app in flash which outputs stunning french voices which indeed sound very real. Worth a mention…havent seen anything like it in flash so far.
Go to the site start editing the video. Add audio.enter some French text. (obviouly the text needs to be french…as the pronounciation is french.
i.e. je ne sais pas. J’habites a Paris.
prendre les jambes à son cou
“…allow machines to not only understand conscious thoughts, but also facial expressions and emotions…”
this seems like a very interesting technology. Not sure how detailed the controller is yet and if really taps into the brainwaves…or rather some muscle groups……in anycase: tapping into the brain will open up an entire different era of HCI. Time will tell I guess.
The new Musicvideo for Radioheads “House of Cards” brings some really Interesting new aspects to Video. Abstract 3D-Music Videos. Very interesting….if this could be merged into a real-time App….the applications could really be stunning….
Also the methods they used to “distort” the data..was interesting…lesson learned: dont do everything digitally….go back and do stuff in the real world…
The first babysteps were made in 2005 with the presentation of my final exam:
The Idea was to create a billboardlike display that would “recognize” a person coming by, and start playing interactive content. That content was supposed to approximately be in line with the “customers” needs. Therefore we would have had to categorize people into target groups in realtime only by a webcam and recognition software. That was, I must admit, at least one step beyond of what me and Marc Isele were able to do in a couple of months. So we decided to focus on achievable things: Basicly realtime Effects like Motion tracking and keying.
We dicided to start our work with Processing, an open source Framework for digital artists based on Java. Our first success was realtime keying. By taking a reference picture we were able to separate “new” pixels from the reference pixels. But one had to be out of the cameras range for a while, and it didn’t work with complex backgrounds. We improved the Algorithm so that a reference image was no longer necessary, but the background problem seems to be a very complex one and couldn’t be solved until now. If anyone has solved that Problem and is willing to commit solutions to the Community, he or she can do it right here!
The next step was to program a solid motion tracking Algorithm. Due to the lack of time we decided to keep it simple: We took one single line of pixels out of the webcam image (exactly in the middle of the image). With the keying function we could mark those pixels that were “active”, so it was easy to calculate the middle of the Silhouette of a person standing and moving in front of the Display. Unfortunately it didn’t work with people below a certain height, for example little children. But for the upcoming presentation that was totally ok. I was studying Communication Design. Presenting a work, based on a moreless self developed software was quite unusual (back then), so nobody there would mock about the rather horrible code we wrote or about some teething problems. Now that we are going public this is different. So please be lenient towards us. We are working on object oriented Versions of I.D.