Microsoft pitching Kinect control tech to telly makers Microsoft has contacted television manufacturers to discuss integrating Kinect's motion control tech into future tellies. That's according to sources "familiar with the subject" who claim the software giant is aggressively pushing Kinect into as many living rooms as possible, The Daily reports. It is also said that Microsoft …
Perhaps they have loads of shipped but unsold stock sitting in warehouses gathering dust. That's what happens when you fudge NPD numbers to create hype. You stuff the chain with product they don't want and hope the resulting hype will eventually sell them all..
However this time it didn't work, consumers very quickly found out Kinect was useless and expensive shit, with serious lag and detection problems and an accute software problem with little more than a handful of jumping and waving games (something that's still not been rectified a year later).
Microsoft really did create a hardware Bob...
errm because it would be extremely useful.
I actually said in a comment months ago they should be doing this as well as pushing it to monitor makers to integrate. The fact that it was one of the biggest selling gadgets of all time and took the T3 Gadget of the Year pretty much puts paid to your statement about it not being wanted.
Fortunately there are quite a few people in the world with considerably more vision than yourself that have seen the possibilities that what was a simple gaming gadget would bring. Even Microsoft have caught on to the fact that it has much much more potential than just gaming and is taking the opportunity that it sees.
Maybe it's time to get the chip off your shoulder about Microsoft and move on with your life.
According to Microsoft's own numbers it would appear that after the holidays sales of Kinect all but dried up. Retailers had ordered more thinking the brisk sales would continue but instead they slowed to a trickle.
So there very well might be warehouses full of Kinects waiting for Microsoft to find a way to unload them. But sales should pick up again now they they are blowing out the 4gb XBox + Kinect for $199.
Yes, licensed from an Israeli company but MS did a lot of the software development as well. MS probably pays a per use license with global marketing rights to the tech. Remember MS spent $500 million marketing the Kinect alone, their no way the Israelis could match that kind of marketing spend. If MS can get this in to say 50 million TVs in 2012 then even a couple of dollars per set license fee is going to give you a billion dollar IPO.
Have you ever tried to navigate XBox menus with Kinect? It's ridiculously slow, inaccurate, and frustrating. MUCH easier to grab a remote and start pressing buttons. Really, what could be easier, faster, and more accurate than pressing buttons? If the idea of replacing your keyboard and mouse with a Kinect sounds stupid then I don't see why replacing a TV remote is any less stupid.
It's frustrating because it's slow and inaccurate, those are engineering problems, the first is piss easy to solve, faster processor, more memory. The second is really a software problem though a better camera will help. Both can be solved if MS has the will and if it doesn't it deserves to become irrelevant. There is a lot of potential in the tech as a control interface if MS has the vision which I think is the biggest question.
XBox menus. There is the problem, the menus are designed for one type of input. Using the Kinect with normal TV menus is a recipe for failure, the interface will have to be designed around the Kinect for it to be more than a poorly thought out gimmick.
It doesn't matter how fast the processor or how good the software, gesture based input is inherently slower and less accurate than pressing a button. When you press a button there is almost no processing required to figure what button was pressed, it happens almost instantaneously. Just making a gesture take much longer than the pressing a button, and then the system still has to process the video input to try and guess what gesture you intended to make.
And it is a guess, it can never be 100% accurate. First it has to figure out what blurry blob in the image is your arm, then track it's motion and then from that try and discern what gesture you intended. It's a process that is subject to all kinds of mistakes. They can make it better but it will never be perfect.
The hand tracking is actually pretty responsive; navigating the UI is slow mainly because you always have to hold poses for several seconds at a time for confirmation. That is not an engineering problem, it's a fundamental flaw with the idea--you can't wave your arms around and expect it to do what you want immediately, but then ignore your arms when you're e.g. reaching for a drink.
As for XBox menus being a problem, no, I'm talking about the menus that were designed from the ground up for the Kinect. And I think they were designed about as well as they could be. It's just that the idea is flawed. The fact that you're talking about the menus as if they were designed for use with the XBox controller makes me think you haven't actually used a Kinect and aren't really in a position to discuss how good/bad it is for these sorts of input tasks.
> gesture-based TV control
Sit rigidly still if you want to keep watching this program. Turn you head to talk to someone and you will get the shopping channel.
> parental restrictions based on facial recognition.
That is just silly. How long before they find that a mask, or face paint, or many other ways defeat that ?
What about spacial awareness? With the main stumbling block for 3DTV is the need to wear glasses which manufacturers are now trying to overcome. (http://www.reghardware.com/2009/09/09/ifa_philips_3dtv/)
Microsoft is also pitching to PC's and improving it's near field of view capabilities, http://www.theregister.co.uk/2011/11/23/kinect_face_applications/ - it's clear it's clever tech with more potential than simple control, i.e. http://www.reghardware.com/2010/12/13/xbox_kinect_air_guitar/
I think it has very good potential for providing the necessary depth of field detection for 3D displays that adjust the parallax barrier to suit the viewers position. I might not be surprised to see it in laptops in place of the regular single camera. It's only a matter of time before they make the system much more compact. Even to provide 3D video recording as well.
I'm not a great fan of microsoft, but they did well to acquire the Kinect technology, just as they seem to have done well with WinPho8.
.................you want to keep watching this program. Turn you head to talk to someone and you will get the shopping channel."
Really? In other words the TV manufacturers are going to integrate Kinect in precisely the way that will make the kit as unpleasant and unusable for the punter as they possibly can because the point you have just made would not occur to them?
"Specific control gesture 1" = Ignore all other gestures from now on unless:
"Specific control gesture 2" = Take notice of all gestures from now on unless:
"Specific control gesture 1"
If I can manage to think of that in two nanoseconds I am sure the likes of Samsung, LG, Sony etc will manage to as well.
I've never used a Kinect, but I thought it had voice recognition built in? Everyone seems to be getting hung up on the gesture side of things, but (assuming what I have heard is correct) surely a telly that you could tell to "switch to Sky One" or "record Coronation Street" would have some appeal?
But so what if it has voice recognition? So do many other things and have had for much longer than Kinnect has existed. If TV makers wanted to use voice recognition they already would since the tech has been out there for a long time.
It's a pretty dumb idea anyway. In most homes people natter away while watching TV. It's annoying enough when you're watching the closing minutes of an important sporting fixture or the denouement of your favourite serial drama and junior walks in and says "I don't want to watch this. I want to switch to CBBC." Imagine how happy you'd be if the TV switched to CBBC.
I think the main issue here is that this is plain yet another system we don't need. Neat idea, but what you want is one person in charge of the TV. In our house when you choose the show you get the remote for the duration of the show. Imagine if the entire family had their own remote.
PS Eye does a similar job but does not have depth perception.
Kinnects main SP is depth perception.
As someone said we need buttons to press.
Move tracks controller position and angles, uses the camera to track in 3D space and gyroscopes to track orientation.
Kinnect tracks the basic body positions in 3D (arms legs head).
This means one is great for sports games where accuracy is important (like a tennis of golf game), or for first person shooters as a gun. But pants as body position games.
Kinnect is very good for dancing games, no so hot for sports games, and forget shooters.
You're assuming TV manufacturers give a crap about making a useable system, as opposed to adding a gimmick to encourage the gullible to buy the new products.
Now I'm not saying they definitely won't do a decent job of integration, it remains to be seen however. There is a definite risk that this may just be used as a half arsed marketing stamp. Like "multimedia PC" which translated as the computer had speakers, and possibly a CD drive.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
