Only Kinect: Microsoft boffins build Minority Report-style tools In the two years since Microsoft launched Kinect, there’s been huge interest among enthusiasts in adapting the hands-free controller to new settings beyond gaming. Techies and geeks of open- and closed-source shades have hacked the hardware, used the tools or adapted Kinect to steer computerised, wheeled armchairs. Among …
I am in no way diminishing the efforts these guys are going to with the AR works. It is nothing short of amazing to be honest, but and it is a big but the kinect is designed to work at an optimal distance and in an optimal space that is bigger than most people have available to them. I know there are lenses that you can buy (3rd party of course) to reduce the focus of the lenses so it works in a smaller space, but I have found that they are not that good.
I have about 2/3 metres between my TV and my couch if I want to use the kinect I have to move the telly around so I can use the length of my living room. I love the kinect in the same way I think that Nintendo were ground breaking with the WII and how the controllers work there. It was a natural evolution to remove the controller. Surely the next evolution would be to enhance the kinect to reduce the space that it is required to work in? As I have already said what they are working on is outstanding, but it seems they are ramping the evolution up a hell of a lot for something that will not be ready for a couple of years when there there is still work to be done for what is out there now. That said keep up the good work, just please spare some people to upgrade the current tech, cheers.
I'm not sure how this relates to the content of the article - experimenting with different ways to use the data from the Kinect is a completely different area to improving the hardware. Making the Kinect able to work in a smaller space with have little or no impact on what they are doing.
I'm sure there are plenty of people working on improving the hardware for the next iteration.
“We are starting to move towards this notion of augment reality. It’s about augmentation with the instrumentation of the environment, instead of augmentation of the user,” he said on a recent visit to Microsoft in Cambridge.
Sounds very much like a Quest for Live Operational Virtual Environments ,which are ARGonaut Domains ..... AIMaster PilotdD Programs with Kinect Server Connects ...... Special Seeds and Fab Fabless Feeds for Cosmic Growth ......... Lovely Creative IT with Leading Virtual Machinery.
The real time bodypart recognition system and clustering algorithm that underpins Kinect's skeleton tracking system was developed in-house at Microsoft Research, as was the infrastructure used to train the recognition system. The early iterations of the GPU implementation of the system was done here too; not sure if the current XBOX code is still based on this mind you. There are even papers out there detailing MSR's contributions.
The software demonstrated in this article was developed at MSR, based on research done in house. There's a chunk of new and interesting stuff coming out in the next K4W SDK (out in the new year, if all goes well) which was also done at MSR.
Without the contribution of MSR, there would be no skeleton tracking system, and without the skeleton tracking system there would have been no Kinect. Without Kinect, do you think that gear from Primesense et al would be so cheap or easily available?
Still building on the initial work of Rare and PrimeSense, though - companies that worked hard to bring a product to market. Without the work of PrimeSense and Rare, MSR would have continued to contemplate their navel, or wasted time on some other concept that went nowhere - they certainly wouldn't have been working on skeleton tracking systems without the work of the aforementioned companies.
This article is worded to suggest that MSR basically invented Kinect single handed, which is rubbish.
The Primesense camera was not the only device considered for the Kinect system, merely the one that was selected as having the best cost-to-performance tradeoff at the time (I worked for one of their competitors at the time, as it happens). MSR built upon Primesense is as much as they used their hardware, but all of the actual hard work of skeleton tracking was entirely their own.
Rare's principle contributions revolved around creating actual products and working software that built upon the recognition system developed by MSR and the MS dev teams. Their higher level design work and product testing helped with the Kinect system as a whole, but the core recognition systems were developed by MSR, not Rare.
As for navel gazing... the purpose of a research lab, either academic or corporate, is to engage in speculative work. It is how you end up with a group of skilled computer vision researchers readily available to deal with issues like tracking skeletons at video frame rate given a data stream from a depth camera; something that no-one else was able to do at the time. Venture capitalists fund many unsucessful startups; research labs generate much uninteresting and useless output. To deride the work of either shows a fundamental misunderstanding of the whole enterprise.
@Zola - As someone who's company has just been purchased by a very large IT company (one of the 5 largest) It is not uncommon for major companies to partner with the company that they're going to purchase for months or years beforehand. It's highly unlikely that MS just bought the kinnect form another company and tooled up a factory. It's much more likely the case that they worked with said company to get from the initial idea to, say, 90% complete before purchasing the company and finishing the remaining 10% (the really hard bit, generally speaking) in house.
To suggest that MS did nothing, sticks of MS hater.
Speaking of which, where are Eulampios, Goat Jam, Eadon and Toadwarrior, I'd like to see how they spin this as MS=teh evils.
I know no-one here seems to like pedants, but this article is just a bit too far. I do try to be relaxed about the internet's generally poor English, but the writer needs to remember that while a good user of English can unpick the errors, those for whom English is a second (or even a first) language that is less well known are really going to struggle with some of the errors as the context can't easily allow the meaning to be understood.
As for the article's subject, I think the biggest let down with Kinect I've experienced is its lack of precision. Hopefully the higher resolution that's supposed to be in Kinect 2.0 will help.
Does that mean Apple would travel back to the garden of eden and slap the Almighty with a patent infringement summons for the forbidden fruit scandal?
That must mean Microsoft are one step closer to patenting a wave, or hand gesture... thin air next?
Paris, IMHO could do with some augmentation... both reality and the other ;)
WRONG - the fact that you have suggested a patent for hand gestures in thin air, means there is prior art. As we all know, the fruity firm are the masters of patenting someone else's ideas. MS just wait for others to invent, then buy the company.
Kinect was quicker, of course.
Of course it was. Microsoft needed an answer to the Wii and the PS Move and they needed it quick. Granted the Kinect does more than simply answer the competition. It's one of the few good innovations (and one of the few truly innovative goods) to come out of Redmond in a long time.
The Kinect did not come out of Redmond. It came out of a university research department. Microsoft licensed a cut down version of the device. LIDAR (which is functionally all the Kinect is) has been around for bloody ages. All the university did was change it from a sweep to a static projection and map distortion in the dot field.
Kinect is brought to you by Microsoft's wallet, not its R&D.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
