The Kinect was a real boon for robotics research. A depth camera that worked pretty well, with skeleton extraction and directional audio? And the price is what!? My lab still uses a first gen Kinect regularly.
Thanks to Jamie Shotton and the team for a sensor that made a difference to an incidental community.
It's still wild to me how consumer hardware outpaced cheap lab tech so decisively around 2010-2012.
At the time I was in a research lab that needed wireless networking + high-def video + light source + on-device processing + battery power + small form factor. It turns out that even compared to Raspberry Pi, the cheapest possible solution to that problem was "buy old Androids on Ebay and add external batteries". A handful of years before that we would have been cobbling together a $500 device, or more realistically scrapping the whole project.
The whole model of feature-dense sensors suddenly crossed from something for major production runs and custom orders into single-unit consumer products, and I'm not sure people really noticed how big an impact it had on research.
This is actually a reason I am kind of surprised to see Microsoft just shut this down outright. Even as a much more limited run product, Kinect has so many possibilities in robotics and research, and it intersects with the VR/AR developments today.
I was talking to someone 3 days ago about how his company has been hoarding Kinects because they knew this day was coming.
I think robotics research is probably a reason why they're discontinuing it. Why support something that only exists outside the walled garden? Maybe not the only reason, but a reason. The device probably isn't generating profits, and the value gain on top of that is missing because the gaming community has abandoned it.
> I think robotics research is probably a reason why they're discontinuing it. Why support something that only exists outside the walled garden?
This is pretty unfair. Microsoft released Kinect 2.0 for Windows[1], with a developer SDK, specifically for the community to hack on. It even supports Windows Embedded.
I can't in my wildest dreams imagine that they thought that product would be a money maker.
Similar deal as Sony removing PS3 features after launch: people found a product they loved, for reasons not envisioned by Sony, who was then hell-bent on shutting it down.
Even though it seems like Sony's interest to foster a community of high-tech nerds interested in AI and parallel processing, and even though it seems like it's Microsoft's interest in fostering a community of robotics and computer vision nerds, it turns out the Giant Corporations need lock-down control on products and have no long-term vision for supporting ecosystems or communities.
The main reason Sony killed OtherOS was that they began to think it could be used to circumvent copy protection for games. In Kinect's case I'm guessing the pricing was set at least in part with the idea that the games would make up for low-to-negative profit margins.
I'm not so sure it was so cut and dried as that. The air force was using them to create a supercomputer using other-OS and Sony was selling it as a loss.
I think they were worried about people buying them up and not purchasing games.
There is the Asus Xtion which is a knock off Primesense Carmine. Google Tango is basically the same technology, as is Apple's Face ID*. Intel's realsense systems are also similar, but I don't know anyone who actually uses them.
Unfortunately Microsoft missed the chance to buy Primsesense before Apple snapped them up. Still, while the tech is good, the ToF sensor in the Kinect v2 is superior for the (gaming) market and Microsoft own the IP via their Canesta acquisition (they licensed the Kinect patent).
What's more interesting is the boatload of cheap ToF systems that you can now buy. The only problem is they're much more power hungry. The new Kinect is crappy for mobile robotics because its heavy (90% of the sensor head is heatsink) and needs a wall power supply. It's fine for static systems, or beefier mechanics though.
(Face ID is literally a mini-Kinect, since it comes out of Primsense IP.)
Occipital still produces the Structure Sensor [1], which uses similar technology to the 1st generation Kinect. They've also recently introduced the Structure Core [2], a new generation, miniaturized depth sensor designed to be embedded into anything that needs that kind of sensing.
I wonder how much the face sensor on the new iPhones is going to cost? Will people be able to buy it as a separate part? I have to think that for repair purposes it will be available, but the price point will no doubt be inflated.
Did I miss something? It's $379 on the site. Was the kinect actually only $18.95-37.90?! Now I wish I had picked one up. I thought they were north of $100.
Most of the development in realsenses has been discontinued a few months ago. We used one (SR300) at the company and discovered that some particular poses (two persons making a kissing motion) would just make the latest SDK (R3) crash, even without feature & gesture recognition active. The answer? "EWONTFIX, use the older SDK, this won't be developed anymore anyways".
Besides, the kinect has honestly better specs, a better quality in tracking and more accuracy.
We're waiting on the 400 series, which have been meant to be out "any day now" for over a year! But the SDK has gone public, and they're saying orders will open soon.
We were very disappointed with the current generation of RealSense devices. The next generation ones look good though - more based on the Tyzx acquisition.
I'm pretty sure cheap and high quality lidar is coming very very quickly - so many companies working on it. There will be no market for an awkward kinect. Kinect is not great compared to modern lidar systems anyhow, it was just more accessible.
Not entirely. LIDAR has other issues important when building self-driving cars. LIDAR can't read signs. LIDAR doesn't work well in the rain, though progress is being made these days.. And don't forget the cameras tesla uses don't rely solely on the visual spectrum we see. They can still see through fog via infrared, for example.
Can't the LIDAR eventually be made to use IR wavelengths and thus synthesize 3-d images through fog?
also, the question of seeing the fog is more nuanced that can see/can't see:
"Just like it is impossible to give a simple answer
to the question “How far can I see with a thermal
imaging camera?”, it is equally impossible to say
how much shorter the range will be in foggy or
rainy conditions. This is not only dependant on the
atmospheric conditions and the type of fog but it is
also dependent on the IR camera used and on the
properties of the target (size, temperature difference
of the target and background, etc)" [http://www.flir.com/uploadedFiles/FOG_techNote_LR.pdf]
LIDAR typically uses lasers in the so-called eye-safe range (around 1.4 micro meters). This range is precisely around the absorption peak of water so that LIDAR cannot damage the eyes of pedestrians and other by-standers. By construction, LIDAR sucks in the fog, rain, snow, etc.
Maybe that was MS' problem. People that buy a 1st gen (2010) and still using it. Not upgrading it. Not replacing it. MS can't possibly profit off of it when only 1 kinect per lab is bought.
That's definitely a huge part of it. But MS would still be happy to lose money on kinect if game makers were making nice games for it.
The only game in my house that could ever utilize it was the original Dancing games from the demos. The last time I saw my daughter playing it, she was holding her phone, and using it for the sensor instead of the kinect through their integrated app.
I asked why she would do that? Just plug in the kinect and use it, don't risk dropping your phone while dancing, but that's not how she wanted to play.
This makes me sad, because my experience with my own Kinect (360) was very much a "this could be awesome if it wasn't burdened with stupid crap".
I bought it for my kids. Turns out, the Kinect is awful for kids. They were far too little for the Kinect to properly see them, and the motions to actually use it were super fussy. Then if I walked in to try to help them, the Kinect would freak out that there was a new person in view. The dog walking by would mess it up as well.
Basically, you would need a large dedicated room for the stupid thing, and you would need to be at least 8 or so to have a chance at using it properly. Then the games were meh.
If you had a large enough dedicated room (I used most of a basement once, with a lot of setup work to give the Kinect the view of just about the whole room) and the right games (some games were less fiddly if multiple people/gestures were recognized that others [1]; though a lot of the management of it is still taking turns and having patience) it was sort of magic to watch particularly young kids play with the Kinect.
Especially then it seemed like a glimpse into a future of where the technology could go, and though that magic was sometimes finicky, it was still magic when it worked.
[1] Kinectimals (essentially a "cat petting" simulator) I recall particularly launching for my youngest cousins to enjoy. There were others, but that's the first to mind; this was a couple years back at a holiday party where most of this happened.
Speaking of magic, universal uses Kinect for Harry Potter in Knockturn Alley. Was surprised to see a simple Kinect running it (hard to see but it’s in a box to the left of the skeleton magic wand experience).
The Kinect has done wonders for amusement park and museum efforts. At this point I'm more surprised when those sorts of museum and amusement projects aren't using a Kinect. (...and more often than that it's simply because an old expensive system hasn't broken yet, but soon as it breaks you expect it to be replaced with a Kinect.)
For that reason, the commodification of the Kinect really has been a boon for science and entertainment. There are some commodity Kinect knock-offs out there (Intel's cameras and sensors come to mind), but selling millions of Kinects means that they will probably stay important to museum projects.
The secondary market will probably remain flooded with Kinects for a while, but maybe (hopefully) by the time it becomes hard to get a decent priced Kinect for such cool little projects there will be a Kinect 3 or similar ready. (Or Microsoft will sell the Kinect brand to a hardware manufacturer.)
I went to a few museums last week on vacation. I was surprised to see how ubiquitous the Kinect was - almost every new-ish exhibit had something interactive that made use of the Kinect. Just when I was thinking about how much fun it would be to get one to play with, they killed it :(
It will be interesting if Microsoft quietly makes a "Kinect" platform that targets businesses / museums / etc. for this exact reason. (Basically, going the Google Glass route of discontinuing a consumer "flop" that still ends up being utilized, and produced, on the B2B end).
Kinectimals was actually well beyond the standard skeletonization system - I think they rewrote a chunk of the stack to allow for younger kids to use it and have it perform well.
I think it was more similar to how Double Fine's Happy Action Theater / Kinect Party handled the system (I recall they had some great write ups of the compromises they found), which came later than my hazy memories of my young cousins but also fit that young of an audience well. The Kinect provided a spectrum of accuracy. The most accurate skeletonizations required the most trade-offs in number of users and unobstructed views, etc, but the Kinect also provided lots of less accurate data, such as the raw depth sensing, and you didn't have to rely on just one data stream. So you might use the depth sensing alone as your primary tool, and there were enough basic gestures you can easily watch that way (like reaching in to "pet a cat" can be a very clear depth change), and use more accurate gestures and skeleton tracking for cases where you actually need more accurate gestures and/or skeleton tracking. You would use one data source to backup the other rather than only programming to the easiest/most accurate/most in-depth model.
I find Kinect programming retrospectives fascinating. Particularly that fascinating feel that with the Kinect you have so much data at your fingerprints that good Kinect programming is as much figuring out (quickly) which data to ignore as which data to use, but also realizing that you don't have to trust just one source of the data you want a gestalt of it. (...and getting a good gestalt to be performant in real time is certainly a challenge.)
> I bought it for my kids. Turns out, the Kinect is awful for kids.
What's sad is that it could have been great, if MS just put a little more dedication and effort towards it. The Kinect Sesame Street games had so much potential. Hopefully another company can do better. Could be wrong but I think Apple bought the company behind the original Kinect's technology
My 2-3 year old kids love the Kinect on the xbone. Well, the love Fruit Ninja on there and have zero problems with it despite their stature. They've played the game weekly for a couple years now.
I just point the Kinect down a little for them (and play on my knees if I join them) and sometimes move one sofa back a couple feet if more than one person is playing at a time.
The improvements in the xbone over 360 in the Kinect department were huge.
I kept waiting for another game in the vein of Fruit Ninja, but none came.
Also use the Kinect for voice controls on movies. It was the equivalent of the Echo but well ahead of time.
One thing that hasn't been mentioned: Kinect was Microsoft's reaction to Wii's success.
I worked at Microsoft when the first Kinect came out. I spoke with a few members of that team (non-engineers). My question was very clear -- are you expecting this to takeover for controller based gaming? I don't recall all the responses, but I think the overall sentiment was along the lines of "No, but it might", whereas my thoughts were along the lines of it absolutely will not, this is such a gimmick. I'm not a gamer at all, but I used to be in college. When I want to game, I plop down on my couch and mash on the controller. If I wanted to jump around and flail my arms, I'd go to the gym or play some pickup basketball.
I think there was (maybe is) a disconnect between Microsoft and hardcore gamers. Kinect and Xbox One's initial non-gaming features were an attempt to take Xbox "mainstream". Stop it. Appeal to the core demographic. To their credit, it seems like they've been doing that now.
> I think the overall sentiment was along the lines of "No, but it might",
This is business though. Lots of stupid side products and features are launched to compete with ideas that might take over. If it doesn't, fine but if it does become popular you aren't 5 years behind and dead in the water.
This is my concern with VR not taking off with the mainstream. Many gamers (myself included) want to plop down on the couch and not move around much. Many of the VR experiences coming out require movement, not to mention the initial effort to put on the headset.
I'm a huge proponent of VR (backer #238 of the Rift on Kickstarter), but I worry VR headsets will end up in the dustbin with the Kinect for this very reason...
vr has larger problems. mainly, the nausea thing: your character can't accelerate at all without giving you really unpleasant nausea. The strength of that nausea isn't being exaggerated. That takes out all games that you would think are good ideas for vr, like sports games, fighter jet games, .. first person shooters where your character walks around.. basically first person anything where your character isn't always moving at constant speed or teleporting is ruled out.
It's a shame.. but it's really how it is
The most popular games are flight simulators and first person shooters where people walk around (Elite Dangerous, Eve: Valkyrie, FSX, Onward, Pavlov, From Other Suns). The best games released, Lone Echo and Echo Arena, take place in zero-gravity environments with highly non-constant speed and no teleportation.
Any game that gets released without joystick/trackpad locomotion will get a flood of angry message board postings.
Nausea is certainly a concern, but huge strides have been made in addressing it for the vast majority of players.
There are seated vr experiences too, but as a Vive owner, I don't think they're going to get cheap enough quickly enough to take off. Maybe we'll get attempt #3 taking off next decade..
VR isn't going to replace flat-panel gaming, it's a new medium, and for more than just games. Think of a VR headset as a new kind of output display, not a new format for video games.
I agree with that. When I say "take off", I mean become popular with or useful for the general public. It makes a great demo though. Also, I'm talking about the current generation of VR that involves strapping something to your face. Get rid of that and everything changes.
Oddly enough, both Wii and Kinect were very successful outside their official gaming applications, because people were able to hack their protocols and use them in ways the manufacturers never intended.
I remember buying a Wii just for the gesture control capabilities. I wrote a whole VJ performance app based on Wii gestures, so you could crossfade video and scrub through animations (the video equivalent of a DJ scratching records), just by waving your hands around. I toured with that thing for years... it was so essential to my performance that I stashed a few candles in my bag, so I could use them for IR tracking in case the sensor bar stopped working.
I then bought a Kinect and used it in a couple performances, projecting 3d mapped effects on to the band onstage. Even with VR/AR tech now: it starts with games but then gets co-opted into art and music. Open, hackable hardware is so important for society.
I bought an Xbox because of the kinect (and two of my friends did the same). I tend to play on my computer otherwise but the kinect was fun and something different. Just wished there had been more games using the kinect better.
Having games that are fun and allow me to play without plopping down in my sofa is definitely a plus. I get enough sitting done when working.
I bought the original Kinect, and I bought an Xbox One after the Kinect was no longer mandatory but I later bought a separate Kinect for it. It has moments of transcendence and a lot of failures between those moments. There's a handful of games where the Kinect sensor really makes sense, and a lot of games where it was shoehorned in and wasn't fun. The biggest problem is that it requires a lot of space. It's a pain to set up, and often a working setup doesn't work for the rest of your life, so you have to rearrange furniture every time you want to use it. It's an interesting piece of tech, but it never really got to be easy enough to use to be what Microsoft wanted it to be.
Another problem is the failure rate and time to failure. Even when we didn't use it often, my unit failed in about 6-8 months. My replacement unit failed in less than 6 months. MS refused to send me a replacement for the 2nd unit. It was a missed opportunity because Kinect was there much earlier than Amazon Echo. MS just wasn't as committed to improving it.
Mine turns itself off and back on every 10-30 minutes of use. Very annoying, but too expensive and too little use to justify replacing out of warranty.
Mine (Xbox One) failed after a year as well. I didn't use the motion sensor as much as the always on Mic. I loved walking into the room and saying "Xbox On" and my Xbox would turn on. "Xbox Pause" and "Xbox resume" when watching a movie and later when Cortana was integrated it became even more useful. But once it failed, I never thought it was important enough for me to replace it.
The lots of space problem is an issue with VR headsets too. Even I have trouble setting up mine because of the space issue and the fact that its just plain easier for me to do something else.
This is a shame. I mean, yeah, it was a pretty awkward gaming accessory, but outside the context of gaming people were doing some pretty fun things with it. A powerful sensor at a ridiculously cheap cost that you could use to create amazing things, provided of course your imagination wasn't weighed down by the latest dumb tech trend. That last statement is just as true now as it was when the Kinect originally came out.
Gives me some strange feels about the current state of the tech economy. The Kinect is being retired at the same time Amazon and Google are caught up in a dumb contest to see who can produce the best hockey-puck-sized speaker that can add items to your shopping list. As overwhelmingly large as tech giants like Amazon, Google, Microsoft, and Apple are, they seem mostly resigned to following trends, as opposed to creating new ones.
Don't forget, smartwatches originated from a highly successful Kickstarter campaign. Same can be said for VR. Bitcoin, and blockchain by extension, originated from a white paper published by someone who's still to this day a complete mystery.
Tech giants are very much capable of generating cash, but are damned by their inability to create gold.
"A powerful sensor at a ridiculously cheap cost that you could use to create amazing things, provided of course your imagination wasn't weighed down by the latest dumb tech trend"
This. And thanks to software/hardware patents, the technology will die with the product, or remain dormant until the owner will either decide to make something else with it or (not holding my breath) release it under a FOSS license.
In the meantime, like always, landfills in the 3rd world will be inundated by more hardware which could be still perfectly useable if its life didn't depend on the wishes of a single company.
> landfills in the 3rd world will be inundated by more hardware which could be still perfectly useable if its life didn't depend on the wishes of a single company.
Well that's a silly comment, because by choosing not to make more then there would actually be fewer kinects in landfills than if they had decided to continue manufacturing. The decision to "build more" has absolutely nothing to do with folks deciding to throw away the ones they've already bought.
I wasn't comparing the pollution of more products vs less products, but rather the pollution of the same number of products after they're forced to become obsolete (years before their technology becomes eventually) by patents and their closed nature which prevents anyone to properly support/use them forcing people to throw them away.
I can even see merit to the argument that cutting off new supply will force people to recycle old supply to repair and keep their sensors running as well. I know I've got a few that I'll have to keep working.
There already is something else “like” it: the front-facing camera in the iPhone X.
On one hand, you could see this as fooling around with Kinect tech now requiring a $1000 investment in an iPhone. (Or buying a Kinect used; there are still plenty on the market.)
On the other hand, you could believe in the inevitability of Shenzhen to take the probably $3 part Apple has designed, stick it into a little housing with a micro controller, and commoditize “Kinect Minis” within the year. :)
While I love Shenzhen's attitude to intellectual property, this is not a solution outside hobby market. I doubt people will dare to release a product in clear violation of Apple's patents (or whoever owns them now).
My expectation is that the no-name hobbyist products (with no clear company to sue) will come first, people will buy them, and then major manufacturers will see them and want to get the hardware into their own products too, at which point they'll probably get interested in licensing Apple's patents. Which means the resulting branded products won't cost $3, but they won't be $1000 either.
I don't expect Apple would willingly license their patent to their competitors in the mobile space (who, after all, would just want to slap a FaceID-equivalent feature into their own phones); but they'd probably be interested in licensing it to e.g. electronic door-lock companies, or action-camera companies (add a "depth" track to your video), etc.
My guess is that rather than the Kinect, what people will probably be buying 2-5 years from now for "commercially scalable" projects is an LG USB webcam that has Apple's miniaturized sensors embedded, and is using the (by-then-standardized) no-name Shenzhen remake of Apple's face-recognition ML-accelerator core.
FYI, front camera in X uses structural light, while Kinect 2 uses a more modern time-of-flight (ToF) technology, which delivers better depth resolution.
"Face ID is enabled by the TrueDepth camera and is simple to set up. It projects and analyzes more than 30,000 invisible dots to create a precise depth map of your face."
Note that the depth resolution they are desscribing is 160x160=30,000, which is really standard structured light. Put a 1Mpix camera behind it and you can interpolate ~6 pixels in between to estimate depth/slope.
Wasn't the Kinect technology acquired by Apple? My impression is the top sensor bar in the iPhone X is essentially a miniaturized Kinect, so we should see more Kinect-like applications now that it fits in your pocket.
AFAIR, it was acquired by ASUS. Original Kinect was developed by an Israel company Xtion, which now lives as a trademark of ASUS game accessories: https://www.asus.com/3D-Sensor/All-Series/
The core technology (structured light in iPhone X and Kinect 1, ToF in Kinect 2) is rather straightforward (for a computer vision researcher, that is...), it's hardware (embedded CPU, cheap sensors and lasers) that enabled its mass adoption.
I don't think the maker community hanging on to their Kinects is going to put a dent in the e-waste impact. People are tossing these things in droves because they are a useless waste of space.
The one _very_ successful aspect of the Kinect was it's speech recognition capabilities. People saying: "Xbox Play Netflix" - that very much preceded the current boom in what you accurately describe as the hockey puck speakers.
Another much under-appreciated aspect of Kinect was what an absolutely phenomenal video conferencing system it was - using Skype on Xbox was on par with expensive custom systems. Great speakerphone (that audio recognition) and it would resize the focus on the fly to include all the faces in the room (panning around).
The tech was extremely impressive, but it always seemed like a classic case of a solution looking for a problem, which to me it's apparent biggest successes as a tool in universities/research groups bore out. Great for Human Computer Interaction type research, everything else not so much it seemed.
At any rate, the idea lives on in Apple's iPhone X in a much more practical application.
> Don't forget, smartwatches originated from a highly successful Kickstarter campaign.
This is a big stretch, in my opinion. I think we'd have worked out that a screen on our wrist with useful info is nice regardless of what happened on Kickstarter, this wasn't exactly an incredible discovery.
There had been countless fitness tracking watches that aren't really all that far removed from the public want from smart watches from Garmin et al long before Pebble tried their thing too. The barrier was arguably getting efficient SoCs that could last a day doing something useful far more than it was anything to do with crowd funding.
> Tech giants are very much capable of generating cash, but are damned by their inability to create gold.
>Amazon and Google are caught up in a dumb contest to see who can produce the best hockey-puck-sized speaker that can add items to your shopping list
But you know why it is so, don't you? The allure for an information company of having an always-on, always internet connected, high-sensitivity microphone, running proprietary software doing god knows what, the allure of having such a device in as many homes as possible is only too obvious.
The only thing more shameful than them pushing this garbage is the consumers' willingness to eat it all up.
Which is why you buy it, and when not at home, play really loud, awkward and uncomfortable recordings of someone having bad, sweaty, cringe-inducing sex. For hours on end.
If you're into software, you could make it exciting. Perhaps write a scraper to strip the audio off porn clips, mix it up with old Khruschev speeches?
Could be a tactic to fight systems like these - overwhelm them with garbage data.
Overwhelming with garbage data is definitely something you can do with social network trackers, it's definitely something I engaged in when I was most active on Facebook. That said it seems to me the far better solution, if you actually value your privacy, is to give up the convenience it would offer and simply not let systems like this into your house.
In a fight to protect your privacy, the only way to win is to not willingly give up your data, even if that means not making use of new systems or taking active countermeasures against tracking systems.
Isn't this the backbone of the entire console gaming industry though? Has there ever been a popular console that directly turned a profit through hardware sales? I was under the impression that basically the entire revenue stream comes from game licensing fees from publishers, with hardware as loss leaders across the board.
Generally they are sold for a loss to begin with early in the generation. But later iterations usually make money. As a whole, successful consoles, usually can make money for the company producing them. That said both PS3 and Xbox 360 lost money for their respective companies. I believe the Wii and the PS2 however were great financial successes though.
AFAIK, all of the recent consoles have been profit making on just the hardware alone. Not very much money per unit at first, but none of them shipped as a loss leader.
> Don't forget, smartwatches originated from a highly successful Kickstarter campaign.
Also don't forget that Sony Ericsson released the MBW-100 smartwatch[1] several years before Kickstarter existed ... so I'm not sure how they could have originated on Kickstarter.
i mean, the amazon echo was a pretty interesting new thing that they created a market for. tthen once they realized people were buying it every big tech company jumped on it to not get behind.
the problem is if google doesn't go into the "speaker with a voice assistant" space, they allow amazon to possibly gain dominance in said space. this is why everyone wants to get in on drone delivery, self-driving cars, vr, etc..
and while "creative", the kinect wasn't a thing microsoft invented either. in fact, primesense, the company that created the original kinect got bought out by apple, and basically that tech is in the iphone x now.
a device that's sitting in the home is priceless to a company. not even for the data they get. it's because the next device they buy they'll think of devices that will work with the one they've already bought. have an amazon echo? they'll probably buy speakers that support alexa, car integrations with alexa, etc..
I used to work at a telemedicine company that uses the Kinect as the primary input for their core product, to enable at-home physical therapy.
Before I left we were doing the work to transition to v2 and were well aware that it was not an earner for MS, but I think the prevailing notion was this was too important to the research community and MS was in it for the long game hoping for it to take off in other industries.
In any case, I reached out to a friend there who tells me they were aware of this eventuality for some time and have another option (smaller, more accurate) that they are in the process of migrating to.
Having invented something is only useful if you're actually able to get that something into peoples hands. As much as I hate how people drool over overpriced Apple hardware, they've done a great job putting greenfield tech into the hands of consumers. Microsoft conversely actually does "invent" a lot of cool things, but for the last 20 years they've been pretty bad at turning those cool things into successful things
"hockey puck that can add things to your shopping list" sounds a whole lot better than "always-on video phone" which they are also competing to put on your shelf.
(and the hockey puck sounds pretty... pretty dumb)
> Amazon and Google are caught up in a dumb contest to see who can produce the best hockey-puck-sized speaker that can add items to your shopping list.
Youre either misrepresenting or woefully misunderstanding this. The echo et al are no more about selling a $50 shopping list speaker than when Amazon/Barnes and Noble/Sony were going at over $50 e-ink readers.
Rolex is a luxury watch maker. The cheapest Rolex you can get is more than 5000 dollars. To say they market to different audiences would be an extreme understatement.
While this is true, it in no way makes my point any less valid.
Pebble went out of business, the very opposite of commercial viability.
Apple went on to outsell a 4.7 billion dollar player in close to a year. If the choice of Rolex offends, insert the name of more or less any other watch manufacturer you like, the comparison usually stands, or at the least shows the exceptional sales the line achieved in a short period of time. If anything, it's even more impressive when one compares to 2016 revenues of cheaper watch makers such as Seiko, Fossil, Citizen etc.
It's not really a shame. The Kinect is out of date and never caught on, and there are much better newer sensors that play the same role for researchers and experimenters. If you'd buy a Kinect for hacking, you'd buy Intel's Euclid or another RealSense camera, e.g.
If anything, it's surprising Microsoft kept building and subsidizing these things for as long as they did.
> Amazon and Google are caught up in a dumb contest to see who can produce the best hockey-puck-sized speaker [...]. As overwhelmingly large as tech giants like Amazon, Google, Microsoft, and Apple are, they seem mostly resigned to following trends, as opposed to creating new ones.
I don't think you're giving these companies adequate credit for their innovation. :-)
It's worth remembering that the Echo, a voice-powered home assistant that can be spoken to anywhere in a room ("far-field" voice recognition), was novel when Amazon introduced it. Similarly, the digital e-ink reader was novel when Amazon launched the Kindle. Two other concepts that seem novel to me are the Amazon Go stores (checkout-less shopping) and Amazon Prime Air (automated drone delivery), both works-in-progress. Consider Amazon Web Services (est. 2006) and the cloud services boom.
Amazon created the first devices/services that resulted in the trends for home assistants and ebook readers (etc.), so I don't think it's fair to name them as "mostly resigned to following trends".
Similarly, although Google seems to have followed Amazon's lead with respect to home assistants, it's certainly innovating in the capabilities of their service, and in other areas. Google's voice assistant has access to incredible amounts of information, presumably powered by Google's search and understanding of questions and context. While both Alexa and Google can answer "When was Abraham Lincoln born?", only Google can more difficult questions like "Who was the second CEO of Microsoft?" (per my test today). However, none of the offerings can yet answer "When was the second CEO of Microsoft born?" Boiling the situation down to shopping lists ignores the capability variation in products.
Of course every company will follow trends to some extent, since to do otherwise is to give up a potential market segment. Companies don't exist per se to create trends; they exist to capitalize on business opportunities. A lot of the big tech companies have expertise they can leverage to enter new tech segments, or have related services that they can integrate. E.g., if Apple has Siri and the Apple TV, then a home assistant may still be a sensible play, even if Amazon did it first.
Trying to create a new trend is risky and requires large investment. For every hit like the Echo, you might strike out like the Fire Phone; for every hit like the Pixel phone you might strike out like Google Glass or have moderate success like Kinect. It makes sense for companies to create offerings when there's a proven model to follow in addition to investing in new innovative ideas.
If following is the only thing a company does, then you might judge them as not being innovative; but before you do that, you should look at the innovative things they're trying too, such as the Glass, the Kindle, the Kinect. Apple has been incredibly innovative in their phone design and security features: the first to offer a touch-based phone with no keyboard (as far as I know); the first to offer fingerprint-based unlock instead of PIN code unlock (so convenient!), and now face-recognition based unlock; Siri may have been the first useful voice assistant. Apple's security and privacy has been industry-leading, e.g. Secure Enclave. Apple CarPlay is a great experience and surpasses every other car integration system I've used.
I believe that the reason all of the named companies have continued to be successful and remain market leaders is because they are innovating in products and services.
Most unfortunate. I was hoping to see the technology applied to VR. One of my complaints/observations about VR is that you are a disembodied viewpoint in the VR environment. Something like a Kinect mocap system would add bodily presence in the VR environment.
>> Something like a Kinect mocap system would add bodily presence in the VR environment.
I personally spearheaded a couple of projects to make it work with the Vive in this way.
Unfortunately, the Kinect's tendency to screw up leg motion, SDK's and API's that read like an old grimoire of Black Magick and worked about just as well, as well as it's proprietary skeleton detection drivers, which prevented me from getting in and fixing a lot of these issues, which caused the company I worked for to drop it completely.
a focus on 'developers developers developers' could've really saved this thing.
There are open-source projects to allow a Vive to leverage Kinect motion tracking. And HTC is developing a full body motion tracker.
These are solutions I found when looking into leg sensors for a Taekwondo VR training game. I feel we're approaching a technology intersection between self-driving cars and VR, in the need to map physical spaces and the objects therein. I keep waiting for someone to bring in eTemplate laser scanning tech to bridge the gap, but nobody has. I fear most people see these as entirely separate domains with little overlap.
I think the technology (basically structured light / ToF) is and will be implemented in many tech, like the Hololens, but also in the iPhone X (Apple bought Primesense which created the first Kinect).
I think we will still see that technology in many products to come.
It is applied to VR. Kinect technology is a part of HoloLens and is used for inside-out tracking in recent crop of Windows VR headsets from Acer, Dell, HP and the likes.
Kinect was a phenomenal piece of technology for a lot of my highly experimental AR stuff.
However, a lack of MacOS support, terrible drivers for Windows that worked about half the time with my Unity rig, non-native Unity support, et al, really messed up a lot of the longer-term plans I had for it, and caused it to not be reliable enough to ever use in a production environment.
The developers make or break a piece of hardware, and while I get that it was mainly an XBOX device, when it failed to make a serious splash there, they could've saved the hardware by working with its high-demand for a ton of different high-tech solutions and provided consistently better SDK's and API's.
Good riddance, because hopefully we'll get something better. Shame, because Microsoft really had a product that spoke to higher-end developers and filled a phenomenal void for a low-cost alternative to brutally high-cost motion capture and natural interaction systems.
Structure Sensor supports OpenNI 2 on Linux, Mac, Windows, and our own Structure SDK on iOS. It oughta do anything you could want from a Kinect and a bit more.
It's a magnificent device for tinkerers. My workplace picked up a few Kinects when we needed a quick & cheap motion capture solution to animate a few characters in our Unity game. We were quite impressed by how well it did; it could even do facial capture! We still had to hand-tune some of the keyframes, but it saved us countless hours of animating.
After getting all of our motion capture done, we wanted to see what else we could do with it. I stumbled across a [1]repo from a few years ago that live-streams the depth data as a point cloud system in the browser. So we cleaned up the code, set up a kinect to point at our office entrance, and now have it live-streaming in the header of our [2]website! It's a bit too heavy on mobile, so we're currently working on a fallback.
We built a commercially viable product with the 2nd generation Kinect that calculates cubic dimensions of packages for the Transportation/Logistics industry. As others have already stated, the sensor capabilities at the price point offered are unbeatable.
For context: our device is capable of measuring the dimensions to a 5mm accuracy on cuboidal (box-shaped) objects sized between 60mm and 1200mm (approx 2.4" to 48")
The device has passed certification by some pretty rigid government bodies in North America that spent a significant amount of time evaluating it to ensure it measures correctly ALL the time.
Of course we always knew this day would come eventually, but it kinda sucks that it had to be today though. Time to buy up those surplus refurbs I suppose..
What many don't know: The exact chips and technology behind the Kinect v1 (as developed by PrimeSense) are still in use today in many products:
We (that is -- shameless plug -- my company DotProduct) combine a Primesense Carmine sensor (basically a smaller version of the Kinect) with an Android tablet and our software to create a self-contained portable 3D scanner (www.dotproduct3d.com).
Occipital has created their own version of the sensor which they call the "Structure" sensor, but it's still based on Kinect/PrimeSense tech (www.structure.io).
Matterport (a YC company) is using the tech in a tripod based scanner to create 3D interior walkthroughs (www.matterport.com).
The reason these sensors are still in use is largely that PrimeSense (at the time they released the Kinect) was technologically about 7 years ahead of every other 3D camera manufacturer in the market. Only very recently we're starting to see sensors that rival the Kinect/Carmine in terms of overall quality.
Doesn't answer your question directly, but I would assume that an example is the new iPhone X face unlock sensor, since PrimeSense was acquired by Apple.
I knew this was coming for the past few months as the supply was dwindling and resellers were selling them for $45 (half of MSRP). Pretty disappointing since I enjoyed working with these for a project. Have to find an alternative now going forward.
"Launched in 2010 with a $500 million marketing campaign"
I am very, very, far from being a MS apologist but I think that dropping Kinect is a really stupid idea, strategically.
MS have dropped quite a few balls recently and this is another one. No doubt in the boardroom this will be an obvious slash to balance the books but it is a bloody stupid idea when you look a bit wider.
This thing was/is a bit of a game changer - not really new, per se, but very useful in quite a few areas. Quietly, many groups in many areas of research and development have used this thing to do things that border on sci fi.
I hope Kinect is dropped somewhere else - it is just good enough to work well.
Kinect is the same thing today that touch screens were in the 80s; touted as a revolution that will take over everywhere, then people realize sitting there waving your arms for yours is not enjoyable, and then it finds a niche and stabilizes at a much smaller size than everyone expected.
When I had an apartment I didn't buy one because I had no room. Now I have a house the xbox one S needed special pricey cables, and there were no games. So I never bought one.
I really like the idea of the Kinect but I couldn't use it when it was big.
Especially as it was a sharp contrast to the history of the Wiimote, where Nintendo had had the opposite reaction and tried to lock DIY makers out once they realized people were using it for unintended purposes.
I'm glad Microsoft was smarter about this, it enabled a lot of otherwise costly to develop ideas to emerge at the time and the gaming use of the Kinect is almost anecdotal in retrospect.
Positioning at introduction is what hobbled the Xbox One -- we mostly forget it now, but while the PS4 was introduced as a gaming device -- all about the games and for gamers -- the Xbox One was introduced as a cable pass-thru device with augmented TV watching, a smart livingroom, an NFL experience, etc. The gaming seemed almost ancillary, and of course classic television was dying (especially among the target market) so a smart TV guide hardly was compelling.
It was just a massive misstep, and was years too late for its focus. They've regrouped but the PS4 has retained the lead from that early head-start.
The misstep for a non-hardcore gamer like me was that they didn't really deliver on the home entertainment hub either. It was a device halfway to nowhere.
The extra $1 it would have cost to include a digital broadcast tuner (vs. a $75 add on; down to $50 these days) would have been a big draw to cord cutters.
Second, the removal of the windows media tools that the 360 had made it less usable as a streaming device.
Felt like design by disjointed committee. I bought it as a media hub (and for the Forza series), but ended up being somewhat abandoned by MS as the focus shifted, understandably, to hardcore gamers.
There's nothing dead about the Xbox One. Yes, the PS4 outsells it. But most games releases are on both systems, and Microsoft comes through with some solid exclusives ever year. I have an Xbox One and a PS4, and with games that come out on both systems, I play those on the Xbox One because Sony's UI is frustrating to me and it just seems like the software is less polished.
XBox is missing several exclusives like Uncharted 4, Horizon: Zero Dawn, and Persona 5 so there are plenty of games on just one system.
IMO, if you have both then it might not seem like a big deal but PS4 seems to have better exclusives and it's consistently selling 2x as much. It's also at 1/3 as many sales as the Xbox 360 so 'dead' may be over rated but it's not in a great place.
I have bought and played all of those games, and they're good games.
2017 has been a better year for PS4 exclusives than Xbox One, although maybe I just feel that way because I like the Forza Horizon games more than the Forza Motorsport games and I suck at Cuphead. There's still plenty of Xbox One exclusives that I'm happy I've played.
Something that has helped those games, and not the Xbox One, is that the exclusivity is just regarding other consoles: All of their big "exclusives" are released on PC too, which allowed people like me to skip buying the console altogether.
This is not necessarily great for Microsoft though, as licenses for third party games bring money, and people like me end up using their online store only when there's no other choice.
I had the 360 and finally broke down a long time ago to get the PS3 for a few games that were exclusive at the time. I was a PS hater for most of my adult console life. Then I started to realize I was paying Microsoft a yearly fee for a system I already bought and PS wasn't asking anything for the same services. The next system I bought was the PS4. There was some FUD about PS4 to start charging the same way XBox does, but alas, I can play my Call of Duty games ONLINE without paying Sony a cent per year. I believe the yearly subscription killed the Xbox.
As a counterpoint, XBOX Live subscription is not too expensive, offers a lot of value through bonuses and freebies, and most large studios are now offering their own subscriptions for users of their games.
Meh, don't want to get into a console war argument, sorry. I'll just say this:
Xbox One is doing poorly compared to PS4 and its exclusives are becoming nothing special. This is especially true this year when many high profile console games are either Sony exclusive or Japanese so they are not on Xbox One.
If Microsoft had launched Xbox One in its current state, it would be neck to neck with PS4. Unfortunately they spent its first year pushing for Kinect, an "entertainment center", and always-online requirements.
Their advertisement summing to "look at all those awesome extra features that you'll never use because they are US-only!" while at same time Sony advertised with "for the players" focus and solid, LOCALISED exclusives didn't do them any favors outside of US either.
Yes, and as of current, and the Xbox One X is the fastest selling pre-order in Xbox history. Each generation has sold faster than then next. I think this is largely attributable (in the XB360 -> XB1; PS3 -> PS4) to the fact that there was such a large gap between generations, so there was pent up demand. Don't mistake "faster selling" with more overall sales by the end of the generation.
As for the PS3 being a bad experience for Sony, that only applies to the first units. By the end of the generation, Sony was almost parity with Microsoft on units sold. So too has Xbox One started to catch up with PS4 (in the U.S.) Microsoft still sells more software per unit though, if I am not mistaken.
Xbox line has always lost MS money (even back in the glory days of the 360) and Xbox One has lagged way behind PS4 since it launched. "Dead" was an exaggeration, I admit. But it has mediocre if not poor sales.
Microsoft has always lumped the Xbox in with other divisions so you can't tell from the financial reports whether it is or has been profitable. The only indication from Microsoft that it is profitable is this statement from Nadella, which is at least a little ambiguous.
Our gaming business now is more than $9 billion and growing profitably.
> "Why press a button to duck, when you can just duck?"
Why sit on your arse and turn a steering wheel when you can just run forty miles to work?
When someone says "Why do X when you can just do Y?" its usually because "just doing Y" is simpler or easier. Turns out that ducking is neither simpler, nor easier.
But it is simpler and easier. You don't have to learn the controls, and actions you know and are familiar with translate exactly into game-space. That sort of game interface can be fun for a certain segment/genre. People still enjoy playing on the Wii. I would concede that its not mainstream, but who knows, maybe we just need to make the right game.
Sad to see this - I recently talked to a startup that was using Kinect for at-home physical therapy. By no means a replacement for PT, but really a huge opportunity for people that didn't have regular/sufficient access to therapy. Hopefully something equivalent can cut through.
I love playing dance games with Kinect. The "Dance Central" series is brutally accurate in its judging. I remember seeing a demo of Project Natal, which was Kinect's codename. It deeply impressed me because it looked like technology from the future.
In fact, I bought an Xbox 360 instead of Xbox One because I wanted to play those specific games and they like other original Kinect titles are only supported on the 360 Platform.
Matterport's camera system, a low cost alternative for laser based systems , for 3d mapping interiors, was originally based on a kinect hack back in 2010. Not sure where they are getting their current sensors, but they're doing quite well, the underlying technology is definitly not dead.
Ah, the Kinect... platform for the first Matterport revision. (Matt B. is so next-level ridiculously creative, Adam Savage would tip his fedora.) Matterport is a really neat mesh/texture capture and stitching platform that I only saw
rough demos run based on multiple fixed CCTVs to do forensic 3D scene reconstruction from UC Davis researchers some years before.
Also, a first-hand story for this category:
Once upon a time, next door to my grandparents house in SJ/Los Gatos region, there lived a renter... not any renter, but a corporate/PhD/engineer whom quit his job to start a startup. After some casual conversation, he mentioned he was working on then something stealth-mode involving machine vision and 3D screen interfacing (think Kinect meets Wii controllers). Many Maybachs, Merc S-series and Aston Martins came by to due-diligence and meet said entrepreneur's startup, team and tech at this rental 3-BD with a dedicated meeting room near Hwy 85. The odd thing was, I never heard what happened to him and has baby. That is until I ran into someone whom did some work for him several years later. (The Valley is a very small place; heck, yours truly toured Woz'es and coworker's homes under construction when I was like 7, but I was more interested in the steep road (Harwood IIRC) out front for high-speed skateboarding/bicycling.)
Anyhow, it turns out, the night before the demo, the CEO made a source code change or something that broke the build. And, there weren't backups because they weren't using a (D)VCS. (Ouch.) I think it was acquired for an "undisclosed sum" for IP and wound down IIRC. What sucks is having to go back to work, kind of a tech "bar closing walk-of-shame," but there's always next time.
Orbbec Astra (https://orbbec3d.com/product-astra/) is still a viable option for applications (such as robotics) that need an RGBD camera. The body-tracking SDK is still in beta at the moment.
Big disappointment. In an age of Echo, and Google Home, I can't understand how Microsoft couldn't make Kinect work.
Also, whatever you may think of Kinect, at least they tried something different. The new XboxOne X is just a spec bump. No attempt at innovating any other aspect of the console.
Did you read the article at all? It says the Kinect sensor lives on in multiple products, including HoloLens and Windows Hello, which is pretty similar to FaceID.
What Apple product/tech are you thinking of? The two off the top of my head are FaceID and ARKit, and both seem quite a stretch to me. I feel like I’m missing something obvious.
Years ago Apple bought the company, PrimeSense, that developed the technology for the first generation Kinect (the one based on IR structured light). From the description Apple gave of FaceID during the keynote I'm pretty certain that FaceID is pretty much Kinect in an iPhone.
Of course the application of it is totally different.
I can see that part, which is why I mentioned FaceID. I just have a hard time seeing it as comparable with respect to the range of the Kinect both in distance and application, as you point out. That’s what motivated my original comment.
Are you aware of anything to the contrary? Definitely interested in learning more if so, given the Kinect is going away. I can also see Apple potentially doing more with this in the future, though they’re not there yet.
It seems more like Microsoft hasn’t been interested in promoting the Kinect (it’s been out for years) or applying the tech elsewhere, and they’re just shutting it down, the timing being coincidental.
I think this speaks to a change in direction to VR/AR systems. It seems pretty clear that long-term tracking solutions will include face/body tracking. Oculus hopes their camera-based tracking can be extended to track full bodies and arbitrary objects. Headsets are getting eye-tracking for foveated rendering.
I'll miss the kinect but it doesn't play nice with newer tracking systems. They confuse each other. That makes it difficult to use in combination with a Vive. I'm guessing they aren't very reliable multi-camera systems either. Unlike the Vive, they don't have a timing mechanism to avoid confusing each other.
I've wanted one for a while for computer vision projects and $40-50 is a fantastic price. Instead I'm looking at maybe getting a 360 Kinect (Certified Refurbished) at $40 and $10 for the usb adapter.
I still love my Kinect. It came out at the end of the 360's lifetime, so I waited for the Xbox One to get one which was sadly a mistake. Games supporting Kinect on Xbox One are limited to say the least.
I am still however enamored with the automatic login. It sees me all the way across the room and logs me in. It's wonderful.
It used to work even better with the first gen controllers which had identifier IR lights in them. My friend and I could swap controllers and continue to play the same side of the screen. It was truly magical. The gen 2 "S" controllers however eliminated the IR lights.
The headset ... Sorry. I'm suspecting/hope that someone will integrate Alexa/google home/HK Invoke with XBox. Should be soooo straightforward. I'd do it myself for a hackathon.
The (updated) article points out that they're still using this stuff in Hololens, Windows Hello, and other gesture-control stuff. I think some of it has made it to the various VR headsets that are coming out for Windows this month.
Also, stopping manufacturing doesn't mean they couldn't restart manufacturing if a demand or an application showed up. The updated article also mentions that the secondary markets are flooded with Kinect v2s for half retail price ($45), which is a pretty good reason to stop manufacturing if the demand isn't there.
It also doesn't preclude Microsoft building a Kinect v3 in the future. (It doesn't even preclude Microsoft building a Kinect v3 in secret in the present.)
The original version worked well on Linux. They then had the skeleton detector and such in a proprietary library. I think there are ways to use the newer versions, but they messed with it all enough to make it rather cumbersome.
This [skeleton data being proprietary] is one of the nails in the coffin for sure.
I get that MS would like to keep developers into it's little Windows / XBOX box, but forcing me to keep a Windows installation for the sole purpose of using your hardware is a great way to get developers looking for an alternative.
I assume they were selling this under cost in order to sell more games (software) which is where they make most of their money. Selling it as a standalone piece of tech would probably require a price hike ultimately making it less appealing and at the end of they day they probably projected they wouldn't move enough units to make it worth it.
I'm sad to see this day arrive, even though the community knew the writing was on the wall. As the first truly affordable, mass market depth sensor, it was a shame it couldn't get a second life as a standalone product.
It also looks like I'm not going to sell any more books. ;)
I've lost track on how many consumer products Microsoft has abandoned over the years. Once they kill their Xbox and Surface lines their Enterprise transition will be complete.
This is quite sad, seeing so many amazing hacked together project that uses the Kinect. I bought a fairly new one at thrift shop for only $15 dollar (!!) a few weeks back.
Sad about the Kinect, but don't interpret this as Microsoft/Xbox failing. They are actually doing a fantastic job with the Xbox brand. The X (scorpio) is launching soon, and they are releasing a bunch of older titles into their ever-expanding backwards compatibility library. In addition, developers are pushing out graphics/performance update patches for the popular titles on the X, and recently overhauled the Xbox UI - it is extremely snappy now and customizable.
Thanks to Jamie Shotton and the team for a sensor that made a difference to an incidental community.
https://www.microsoft.com/en-us/research/video/body-part-rec...