Slashdot Mirror
Microsoft Shows Off Future Product Tech
adeelarshad82 writes "Microsoft opened a portion of its fifth TechFair to Silicon Valley residents, demonstrating more than 15 technologies, which included everything from real-time translation to mobile-to-mobile networking to improved image stitching. The top two that really stood out were the translating telephone, which actually used no 'telephone' at all — it was a test to discover how well Microsoft's speech algorithms could interpret speech, translate it, and then speak the translation using text-to-speech algorithms — and Manual Deskterity, a new paradigm for a user interface; a right-handed user's left hand, for example, can be used for coarse manipulations of objects, while the right can be used for fine manipulation, such as with a pen. It sounds a bit simplistic, at least at this stage. Since one of the charters of Microsoft Research is that the work should eventually be moved to product teams, there's a good chance that the prototypes will eventually be made available to the public at large."
Two words: Courier Tablet
I judt got a nre Kinesis keybiartf so please excusr ant egregiou typos.
One of the technologies showcased looked a little bit interesting. It was a "pseudo-3D". Here's from TFA:
"Akeley's prototype uses depth filtering, layering different focal planes on top of one another to give the eye something to focus upon as a 3d object moves "closer" and "farther" from the eye."
Could it mean 3D without funny glasses? I have no idea, but I hope so.
You are welcome on my lawn.
Will any of this new MS technology help Slashdot process Javascript better?
New
Innovative
Revolutionary
Creative
Cool
Microsoft
Can you figure out which one?
If I were God, wouldn't I protect my churches from acts of me?
Quite a few actually. See http://research.microsoft.com/en-us/
This might not be far-fetched at all. I identified decades ago, before I was even legally an adult, that I had precisely that division of labor between my two hands and arms, and indeed probably between both entire halves of my body and brain: one half performed brute force maneuvers requiring strength, while the other specialized in performing actions requiring precision. Thus I write and manipulate a spoon with one hand, while using the other to throw a ball, swing a bat, and wield a steak knife. To some degree I've tried to thwart this as an adult, by trying to "teach" each hemisphere to be less specialized, but I still throw like a girl with that other hand.
Browsing through there I found one - Destination Maps.
Only it doesn't show up at all in Safari using Bing Maps. Have they really delivered it? I cannot say.
They have a ton of talk about the amazing integration of Microsoft R&D in products, but again very little I see as shipping things or core features - even Destination Maps is more of an adornment than a core feature, and one I doubt more than a handful of people use day to day.
The crime of it all is they do have brilliant people there, but they up doing the world very little good.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
Microsoft seems to have no aversion to putting out beta-quality software -- why not publish some more interesting betas?
Market testing.... That's what MS does with new ideas... but it doesn't mean they will produce.
I wish I could be more impressed by that.. but I'm not. It's already trivial to...
1. take a video source
2. split up into images
3. from those images, sort by quality (least blur to most blur)
4. map them onto the panoramic plane (thanks to it being video, you can use motion vectors to help this process, but it's not really needed.. existing algorithms will chew through them easily enough)
5. remove those images that are superfluous (i.e. not needed for covering the entire canvas)
6. blend (using al algorithm a la smartblend to try to keep the number of decapitated people down)
You can have fun with some movies that way...
Terminator 4: http://img194.imageshack.us/img194/1629/moviepanosterminator4.jpg
The Mist (may be considered a spoiler, for those who haven't seen it yet): http://img41.imageshack.us/img41/3221/moviepanosthemist.jpg
The article notes that the panorama of one of the microsoft buildings is 'crisp'.. well sure, when scaled down a good fraction of the original video's size in terms of coverage. Now if they combined multiple frames for a superresolution image (similar to how astronomers might image stack a bunch of blurry shots and out comes a high-resolution detailed image) and thus panorama, that'd be more fun.
"Manual Deskterity" (did not see this term referenced in the slideshow by the way - where did it come from?) sounds like something they developed just so they could patent it. This seems like the natural progression for something like the Wiimote or possibly even an iPad-like device, considering it's what people do naturally anyway.
Funny thing I noticed flipping through the slideshow: many of the annoying pop-up mouseover ads open a bing window (including the word "Apple"). Of course there are also Google ads unobtrusively placed under the main content of the page. If only Microsoft had some people around who could research ways of making their ads less annoying...
Microsoft opened a portion of its fifth TechFair to Silicon Valley residents, demonstrating more than 15 technologies...
Good god, I thought schools taught people to count up to one hundred AT LEAST.
Or you can just plug a second mouse in. People freak out when they see it ("that can't work!") but it does. Just pick two mice with a different number of "mickeys" resolution.
It would be nice to see a real article about the more interesting sounding things but as it is, I'm not impressed.
Check out this page, Paul Haeberli's wonderful old site Grafica Obscura from when he was at SGI.
http://www.graficaobscura.com/merge/index.html
This is his famous image merging by projective warp program, where he could take a bunch of snapshots and automatically warp and stitch them together. I think this is from 1995 or so. It references papers from 1991 and this one from 1994: S. Mann and R. W. Picard. Virtual Bellows: Constructing High Quality Stills from Video. IEEE International Conference on Image Processing, Mov, 1994.
Many researchers have published about extracting features from blurry views. I remember one ACM Siggraph article I believe from years back about using multiple beams to achieve high resolution images underwater for example.
The deskterity and pseudo-3d sound interesting but honestly there is very little to see in this article.
A few months back, we were seeing comparisons between five Microsoft mice prototypes against the then-new Apple Magic Mouse (the mouse with a build in touch surface). What everyone recognized then and seems to recognize now is that there's a big difference between an actual shipping product and prototypes that may be cost prohibitive, cumbersome, or otherwise unacceptable to consumers due to the numerous details that need to be worked out in the process of going from prototype to product. Microsoft is quickly becoming one of the biggest makers of vaporware in the market, and they need to stop it before they lose (even more) credibility.