Beta

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Kinect's AI Breakthrough Explained

Soulskill posted more than 3 years ago | from the expensive-hacker-toys dept.

AI 97

mikejuk writes "Microsoft Research has just published a scientific paper (PDF) and a video showing how the Kinect body tracking algorithm works — it's almost as impressive as some of the uses the Kinect has been put to. This article summarizes how Kinect does it. Quoting: '... What the team did next was to train a type of classifier called a decision forest, i.e. a collection of decision trees. Each tree was trained on a set of features on depth images that were pre-labeled with the target body parts. That is, the decision trees were modified until they gave the correct classification for a particular body part across the test set of images. Training just three trees using 1 million test images took about a day using a 1000-core cluster.'"

cancel ×

97 comments

Sorry! There are no comments related to the filter you selected.

1000-core cluster? (-1)

Anonymous Coward | more than 3 years ago | (#35625056)

Where can I buy one of those? Kinect is amazing and all but not having to wait 2-minutes for windows to load would be even more incredible! It's funny, computers get more and more powerful but yet windows still takes the same amount of time to load. lulz

Re:1000-core cluster? (1)

Luckyo (1726890) | more than 3 years ago | (#35625080)

Any decent large data center will be happy to rent you one for a price?

Re:1000-core cluster? (2)

davester666 (731373) | more than 3 years ago | (#35625270)

Why would MS rent/buy processor time? They've got the world's biggest botnet, and they even have the suckers pay MS to join it.

Re:1000-core cluster? (1)

metalmaster (1005171) | more than 3 years ago | (#35625112)

Wouldnt you still ned software capable of using all of the resources?

Re:1000-core cluster? (1)

woolpert (1442969) | more than 3 years ago | (#35625440)

Forget the 1000-core cluster. I want to know where I can get 1,000,000 images of people with all the (major) body parts zoned and referenced.

That's an impressive test corpus.

Re:1000-core cluster? (1)

lrnj (1986582) | more than 3 years ago | (#35626670)

I would assume they just used an established motion tracking system in parallel with the Kinect sensor input.

At 30 fps, that's about 10 hours of input.

Re:1000-core cluster? (0)

Anonymous Coward | more than 3 years ago | (#35628404)

[Neo voice]we need porn. A lot of porn.[/Neo voice]

First Post! (-1)

Anonymous Coward | more than 3 years ago | (#35625064)

First Post!

This is actually pretty brilliant.

liars, touts & shills oh my, babys still dying (-1)

Anonymous Coward | more than 3 years ago | (#35625124)

cnn emulates real life for US? babys et al differ

hard to tell which smells worse? the fog of tax free (for some) war can do that? did we say tax free? pardon, the non-taxpayers actually profit ($billionerrors$) on the heavy weapon (keeping ALL sides supplied including mexico) murder massacre business outings. so that's good?

we support the views of this former person
http://www.youtube.com/watch?v=TY2DKzastu8&NR=1&feature=fvwp ("stop killing")

we do not support the material in this cnn propaganda video from yesterday
http://www.youtube.com/watch?v=BXB75IK6pL4 ("we can win this, with my help")

same guy? clone? confused? schizophrenic? we must focus... on the images. we must....saw a picture of one of those godaffy psycho-killer freaks being paraded around our military bases (may have paid for them, along with our holycost tithing's) like royalty, only to become our very worst 'enemy' just weaks/leaks later? focus-pocus?

babys rule, with tiny chubby soft fingers, advanced dna etc..... unclear?

real math; taking one (1) life =crime vs. humanity

give US a minute here. this can't be right? isn't there justdenyable homicide? that's the old time religion? god's will? too many of us (by about 5 billion)? still foggy? in these complex times, it can be disgustingly enlightening to return to the teachings of the georgia stone trustdead freemason 'math'?

freemason kids traumatized by native teachings

we're not the only (chosen) ones? the natives must have made some mathematical errors? let's see, wasn't that problem taken care of before? & before that. let's check the georgia stone, all the answers are there? not to fret then, the #s never lie?

the GSM get their tiny (ie; selfish, stingy, eugenatic, fake math) .5
billion remaining population, & the money/weapons/vaccine/deception/fake
'weather' alchemist/genetically altered nazi mutant goon exchangers, get
us? yikes

the 'fog' is lifting? more chariots will be needed?

with real math, even being remotely involved in lifetaking (paying for, supplying endless ordinance) is also a crime against ALL of the world.

ALL (uninfactdead) MOMMYS......

the georgia stone remains uneditable? gad zooks. are there no chisels?

previous math discardead; 1+1 extrapolated (Score:mynutswon; no such thing as one too many here)

deepends on how you interpret it. georgia stone freemason 'math'; the
variables & totals are objective oriented; oranges: 1+1= not enough,
somebody's gotta die. people; 1+1=2, until you get to .5 billion, then
1+1=2 too many, or, unless, & this is what always happens, they breed
uncontrolled, naturally (like monkeys), then, 1+1=could easily result in
millions of non-approved, hoardsplitting spawn. see the dilemma? can
'math', or man'kind' stand even one more League of Smelly Infants being
born?

there are alternative equations being proffered. the deities (god, allah,
yahweh, buddha, & all their supporting castes) state in their manuals that
we needn't trouble ourselves with thinning the population, or being so
afraid as to need to hoard stuff/steal everything. chosen people? chosen
for what? to live instead of us? in the case of life, more is always
better. unassailable perfect math. see you at the play-dates, georgia
stone editing(s) etc... babys rule.

exploding babys; corepirate nazis to be caged (Score:mynutswon; hanging is too good for them?)

there are plans to put them, (the genetically, surgically & chemically
altered coreprate nazi mutant fear/death mongerers (aka47; eugenatics,
weapons peddlers, kings/minions, adrians, freemasons etc...)) on display
in glass cages, around the world, so that we can remember not to forget...
again, what can happen, based on greed/fear/ego stoking deception.

viewing/feeding will be rationed based on how many more of the creators'
innocents are damaged, or have to be brought home (& they DO have another
one) prematurely.

Re:liars, touts & shills oh my, babys still dy (-1)

Anonymous Coward | more than 3 years ago | (#35625248)

I want to shit on your chest as I cum in your mouth.

Re:liars, touts & shills oh my, babys still dy (0)

Anonymous Coward | more than 3 years ago | (#35626266)

IP (#35625124) resolved to slashdot user Wovel (964431).

More advertising masquerading as news (-1)

Anonymous Coward | more than 3 years ago | (#35625132)

It's in the subject line

Re:More advertising masquerading as news (4, Informative)

symes (835608) | more than 3 years ago | (#35625190)

I don't think so this time. This is a reasonably well written formal paper sent for peer review. It is also quite nice to see this published openly.

Re:More advertising masquerading as news (2)

Raenex (947668) | more than 3 years ago | (#35626012)

It is also quite nice to see this published openly.

And no doubt backed up by a dozen patents.

Re:More advertising masquerading as news (3, Insightful)

Jeremi (14640) | more than 3 years ago | (#35626968)

And no doubt backed up by a dozen patents.

Of course. That's the purpose of patents, to encourage inventors to publish their inventions openly.

Re:More advertising masquerading as news (1)

Raenex (947668) | more than 3 years ago | (#35628204)

I'd rather they kept their secrets and let somebody else figure it out than be granted a monopoly on an idea.

Re:More advertising masquerading as news (1)

X0563511 (793323) | more than 3 years ago | (#35640006)

Ah, so you want shorter patent terms and non-ridiculous licensing costs.

Yell at the government regarding the former, and yell at the sellers regarding the latter.

Re:More advertising masquerading as news (0)

Anonymous Coward | more than 3 years ago | (#35625332)

Yeah, amazing how any post about anything some neckbeard doesn't like because it's MS-related is shilling or advertising. Douche.

Re:More advertising masquerading as news (0)

Anonymous Coward | more than 3 years ago | (#35626950)

And here I am again, Mr Anonymous Coward, talking to myself.

Yes I am a douche.

Sounds like vision, all right (4, Interesting)

liquiddark (719647) | more than 3 years ago | (#35625138)

Layered classification nets have always struck me as the right approach, particularly as we learn more about how human senses work - it seems like a lot of our "thinking" is done much closer to our sense organce than we might have once imagined. Interesting that the less "organic" type, decision trees, were used rather than neural nets. One wonders if maybe it was more a matter of ease of phrasing/training/debugging than of classification itself that decided which type to use.

Re:Sounds like vision, all right (4, Insightful)

hoytak (1148181) | more than 3 years ago | (#35625328)

Random forests have always been a nice classifier to use when working with really wacky data types. This is due in part to how easy it is to customize them; a lot of the ways they can be tweaked and tuned and customized have fairly intuitive effects on the outcome and behavior of the classifier. In my experience, while neural nets can also be pretty powerful, they are often much harder to work with as the parameters you have for tweaking can be really non-intuitive. We sometimes joke about neural nets being "black magic" because the training and tweaking can be really uninterpretable.

However, the biggest reason random forests were used is probably because they are extremely fast on current chips, probably a couple orders of magnitude faster than neural nets when the trees are hard coded.

Re:Sounds like vision, all right (-1)

Anonymous Coward | more than 3 years ago | (#35626064)

The real breakthrough will be in random segments of code, grouping together to form unexpected protocols.

Re:Sounds like vision, all right (4, Interesting)

Game_Ender (815505) | more than 3 years ago | (#35625370)

Yep, it's not exactly an AI break through but it's really cool to see a practical application of machine learning in the consumer arena.

Re:Sounds like vision, all right (4, Interesting)

Twinbee (767046) | more than 3 years ago | (#35626682)

Yes, now all they need to do is fix the lag which can be quite high, maybe even 200ms:
http://www.youtube.com/watch?v=weZOjotbuSU [youtube.com]

Something really low like 16ms or better is needed so that we don't notice, according to this article:
http://www.sussex.ac.uk/Users/km3/hfes.pdf [sussex.ac.uk]

Re:Sounds like vision, all right (1)

amorsen (7485) | more than 3 years ago | (#35628842)

The youtube video doesn't really prove anything. The lag could just as easily be introduced by the TV or the game.

Re:Sounds like vision, all right (1)

dominious (1077089) | more than 3 years ago | (#35628392)

Yep, it's not exactly an AI break through but it's really cool to see a practical application of machine learning in the consumer arena.

I suppose that's what the title "AI Breakthrough" means. Training decision trees in a random forest is not a breakthrough.

Re:Sounds like vision, all right (2)

hvm2hvm (1208954) | more than 3 years ago | (#35628528)

Exactly... sometimes "good enough" is better than "it should work in theory but we don't have the required hardware/algorithmical/whatever capabilities yet". It probably won't work perfectly in some cases but for most applications it's great.

I really think AI will be created in the same way. Once in a while a need appears for a AI related task and someone finds a "good enough solution". In time, someone will need a robot to have a serious conversation with and there will be enough knowledge lying around that it will be easy to create that "good enough" solution.

It will be like with the kinect and wii, no-one will expect what will come out of it but everyone will think "hm, they should have done that years ago".

Re:Sounds like vision, all right (1)

oliverthered (187439) | more than 3 years ago | (#35626566)

Go search for Women Aspergers interview tony attwood.

Keep listening until you get to the 'sixth sense' bit.

You may not realize it, that doesn't mean that other people aren't 100% aware.. (e.g. I'm in the third person, it's pretty apparent that I don't make the spelling mistakes I just tell my body by pushing a command out to write some stuff, and it cocks up sometimes).

It does similar in the other direction, with various levels of indirection... and I can also push things further down for some real number crunching.

1000 core cluster haha (0)

Anonymous Coward | more than 3 years ago | (#35625178)

I have more than 1000 cores in my desktop lol maybe they just hooked their desktops together

Re:1000 core cluster haha (0)

Anonymous Coward | more than 3 years ago | (#35654754)

I hooked up to your mom's core last night heyooooooo

ANN? (1)

Gulah (1983618) | more than 3 years ago | (#35625188)

Smells like Neural Networks thinking ...

Strange Descriptions... (5, Funny)

Anonymous Coward | more than 3 years ago | (#35625234)

- "What do you do for a living?"

- "I train trees to make a decision forest that can see human limbs."

- "Ah, I see. Makes sense. (WHAT THE FUCK???)"

Re:Strange Descriptions... (0)

Beelzebud (1361137) | more than 3 years ago | (#35625472)

LOL Someone with points mod this up!

Re:Strange Descriptions... (-1, Offtopic)

93 Escort Wagon (326346) | more than 3 years ago | (#35626358)

LOL Someone with points mod this up!

No.

Re:Strange Descriptions... (0)

vgerclover (1186893) | more than 3 years ago | (#35626888)

Don't worry, I did! :D

Wait...

Re:Strange Descriptions... (1)

Slutticus (1237534) | more than 3 years ago | (#35626918)

Sounds like an upcoming xkcd strip.

Re:Strange Descriptions... (0)

Anonymous Coward | more than 3 years ago | (#35629706)

- "What do you do for a living?"

- "I train trees to make a decision forest that can see human limbs."

- "Ah, I see. Makes sense. Who do you work for?"

- "Microsoft."

- "(WHAT THE FUCK???)"

[Fixed that for you]

Re:Strange Descriptions... (1)

Sal Zeta (929250) | more than 3 years ago | (#35630166)

-"Oh! So, you're the one who writes lyrics for Radiohead, then."

Need a more descriptive summary (1)

radarsat1 (786772) | more than 3 years ago | (#35625276)

From the summary it looks like they are basically using a classifier which they spent a lot of time training, and it works well. This is impressive, but I don't know if it meets the story title's claim of "AI breakthrough", since from the summary it sounds basically like, "researchers used classifier for classifying data and it worked!" Can someone summarize in a little more detail exactly what the "breakthrough" entails, other than basically standard use of classifiers for training on data sets?

Re:Need a more descriptive summary (0)

Anonymous Coward | more than 3 years ago | (#35625334)

Can someone summarize in a little more detail exactly what the "breakthrough" entails, other than basically standard use of classifiers for training on data sets?

TFA says "it is all based on fairly standard classical pattern recognition"

Re:Need a more descriptive summary (1)

narcc (412956) | more than 3 years ago | (#35625486)

TFA says "it is all based on fairly standard classical pattern recognition"

I'm a science reporter. I just want to clarify your above statement -- Are you saying that this is an unprecedented breakthrough in artificial intelligence research that will lead to "thinking machines" in the next year?

Re:Need a more descriptive summary (1, Informative)

mikael (484) | more than 3 years ago | (#35626096)

The function: f=d(x+u/d(x))-d(x+v/d(x)) would calculate the depth gradient of the pixel. It's possible to reconstruct a three dimensional shape from a 2D image [weizmann.ac.il] .

Then your problem is trying to match a human skeleton to the shape. If you know the curvature of the gradient at a particular point, you can eliminate some body parts. A head is mostly spherical and within a particular maximum/minimum, limbs and the torso are more cylindrical with a linear depth along one axis. Look for that linearity, and you could determine that is a limb and what direction it is aligned in.

Re:Need a more descriptive summary (2)

marcansoft (727665) | more than 3 years ago | (#35628654)

This has nothing to do with reconstructing a depth image from a 2D image. The Kinect is a depth camera and already gives you a real depth image (not a guess).

Focussing on the normal bit (1)

Anonymous Coward | more than 3 years ago | (#35625282)

So they fed an LCS with some sample data? OK, par-for-the-course. I'm far more interested in how they generated those '1 million' pre-labelled test images in the first place.

Re:Focussing on the normal bit (2)

hedwards (940851) | more than 3 years ago | (#35625356)

The same way that cybercriminals crack captchas, they just offered up a picture of a random boob to a random boob. The real problem was stopping at 1m pictures.

Re:Focussing on the normal bit (2)

multipartmixed (163409) | more than 3 years ago | (#35625374)

> I'm far more interested in how they generated those '1 million'
> pre-labelled test images in the first place.

Snapshots from the webcams attached to computers running Windows.

Re:Focussing on the normal bit (4, Informative)

gmaslov (1983830) | more than 3 years ago | (#35626404)

So they fed an LCS with some sample data? OK, par-for-the-course. I'm far more interested in how they generated those '1 million' pre-labelled test images in the first place.

I read the paper; it was clever. They used a standard motion capture setup with their actor(s) going through several hundred different movements. Since their algorithm is stateless, they could analyze the motion and produce many distinct poses from each movement. Each pose was then "retargeted" (a well known technique in animation; example [snu.ac.kr] ) onto many different 3D models of people of varying height, body type, etc., before finally being rendered into a perfectly labeled depth map.

They went through several iterations of this process:

  1. Train their algorithm on this huge data set
  2. Notice that it doesn't work so well in some situations
  3. Have their mo-cap actor(s) produce additional data to cover those situations
  4. Process the new mo-cap data into however many thousands of additional training poses
  5. GOTO 10

Re:Focussing on the normal bit (1)

im_thatoneguy (819432) | more than 3 years ago | (#35634148)

What I find really interesting about this approach is that it's machine learning in a virtual environment.

They essentially taught a game controller how to be a game controller by feeding it virtual players inside of a game.

I suspect this is how we'll want to train all artificial intelligence agents. Why go through the trouble of building a robotic body for an AI to use when it can simply be provided a virtual world to live and grow in.

I've also always been curious why more AI research doesn't take place in virtual environments. A prime example is the DARPA driving challenges. You could simulate LIDAR data and stereo camera arrays in a quite photorealisitc environment 24/7. Taking it out on the road seems like a formality.

"Almost as impressive"? (1)

Anonymous Coward | more than 3 years ago | (#35625364)

Ummm, all I've seen so far apart from this are pretty obvious uses of the depth sensor.

What Microsoft has done is solved an extremely hard AI problem. Check out the body-part identification. I think more credit is due.

Re:"Almost as impressive"? (2, Insightful)

Anonymous Coward | more than 3 years ago | (#35626086)

Hum, no, actually, they just used a known for years technic of machine learning on a huge sample of data and it worked pretty well.
From my point of view, there is no major breakthrough but still it's a nice solution.

Re:"Almost as impressive"? (-1)

Anonymous Coward | more than 3 years ago | (#35628420)

My point is what Microsoft has done with the Kinect is much more impressive than what anyone else has done with it, at least from what I've seen.

Also, almost every applied AI paper will use a 'known technique'. What matters is the details of the implementation.

Impressive. (4, Funny)

Chocolate Teapot (639869) | more than 3 years ago | (#35625408)

Training just three trees using 1 million test images took about a day using a 1000-core cluster

Trees have traditionally been trained in Entish, which although reliable, is such an un-hasty language.

Re:Impressive. (-1)

Anonymous Coward | more than 3 years ago | (#35625484)

Very impressive, but you are not a Jedi yet.

Seriously though, I can has source and relevant data files? Please? Just look at the awesome that has been done based on hacked-together open source drivers alone. Open the classification routines. It won't hurt - if anything, it will help drive the sales of even more Kinect cameras.

Re:Impressive. (1)

feedayeen (1322473) | more than 3 years ago | (#35625504)

Training just three trees using 1 million test images took about a day using a 1000-core cluster

Trees have traditionally been trained in Entish, which although reliable, is such an un-hasty language.

The 1K core cluster is mostly because it takes such a long time to say anything. Had they gone with a one core cluster, by days end, the system will have just managed to say 'good morning'. The end result is that it has accomplished nothing. Thankfully, with this system, they can complete this statement in a thousandth of the time, in other words, it reduced the startup time to 28.8 seconds.

Re:Impressive. (1)

TheoMurpse (729043) | more than 3 years ago | (#35627004)

640K ought to be enough chlorophyll for anyone.

Very impressive (1, Flamebait)

artor3 (1344997) | more than 3 years ago | (#35625430)

A lot of the MS-haters on Slashdot tried to write off the Kinect as a nice bit of third-party hardware with a crappy MS-made driver. I wonder how they'll respond to this. Microsoft has really outdone themselves here. I think Penny Arcade [penny-arcade.com] put it best. If only they could apply this sort of innovation to their more important products, they'd be back on top in no time.

Re:Very impressive (0)

Anonymous Coward | more than 3 years ago | (#35625830)

Everyone who knows about it already knew they made a huge-ass prototype like this for testing it.
Then they took the results and created the hardware for it. Then optimized the hardware several times to see how far they could go before it was gimped too much.

Sadly, they went too far and it tends to mess up quite often, actually. And that is in perfectly fine, open, and ambient lit rooms from a 3rd floor.
It really sucks for things fast action.

As for the device and things like that in general, just like Eyetoy, these will never, ever replace a controller. Ever.
The interface is too limiting. It cuts out a HUGE number of possible things you can do with it. It removes a significantly large amount of precision controllers do.
This is the main reason people hate on it.
Move, Wii, even the wand MS made a while back for PC, all more useful.
The whole item-scanning thing they went on about sounded interesting. Has that went anywhere?
The image-recognition for identifying fixed objects you need is pretty simple, such as scanning for a skateboard as one of their examples, simple edge detection algorithm would get it >90% of the time, depending on how precise you made it.

Re:Very impressive (1)

Dr Max (1696200) | more than 3 years ago | (#35626082)

I for one welcome the day i no longer have to push analogue sticks around or furiously slide a mouse around a desktop to take out a guy in the latest fps. Give me a good head and gun tracking system and maybe a heads up display any time. I agree we aren't at nirvana yet but it wont be long before your controlling your rts armies with hand signs from the heavens.

Re:Very impressive (1)

drinkypoo (153816) | more than 3 years ago | (#35626332)

As for the device and things like that in general, just like Eyetoy, these will never, ever replace a controller. Ever.

I think it's fairly clear that the future involves both approaches, sometimes both in one game. Keeping gamepad support anywhere it is possible to do so keeps the game accessible for as many people as possible, e.g. the disabled. But I really enjoy the fact that the Wii gets me moving around. I imagine I'd enjoy the same thing about Kinect (but my 360's optical drive died and I have been extremely lazy about replacing it. I have all the pieces...)

Re:Very impressive (0)

Anonymous Coward | more than 3 years ago | (#35626438)

This is slashdot after all. That being said I think at the time the Nintendo wii was fairly new so a lot of folks (myself included) suspected that microsoft was going to arrive late into the game and put out an "ok" product.

But I'll give them credit. The Kinect and it's homebrew scene are seriously making me consider buying a MS hardware product for the first time. Some of the things they're doing with the Kinect now is really interesting.

Re:Very impressive (0)

Anonymous Coward | more than 3 years ago | (#35626608)

What? Has Penny Arcade a contract with Evil Corporation, by chance?

First, that "current" (2002, I think) strip about M$, now depicting Kinect as a divine gift. I bet it can be used as sex toy, too...

And BTW, just so you know, I don't like being classified as MS-hater... it's M$-hater, mmmkay?

Re:Very impressive (1)

Concerned Onlooker (473481) | more than 3 years ago | (#35626614)

Well, I really like my Mac and I think this is cool. So, stuff that data point in your decision forest. :-)

Re:Very impressive (0)

Anonymous Coward | more than 3 years ago | (#35626650)

Well, this is because Microsoft Research is NOT microsoft.

It is a research lab funded by microsoft doing lots of interesting stuff that does not necessarily mean that it will be introduced in real Microsoft products

Other labs in the past include
Xerox PARC : remember the GUI ?
IBM's BOCA RATON : created the first PC

Re:Very impressive (1)

wjsteele (255130) | more than 3 years ago | (#35626886)

"IBM's BOCA RATON : created the first PC"

And here all this time I thought the Apple computer was out before the IBM... silly me.

Bill

Re:Very impressive (1)

SpinyNorman (33776) | more than 3 years ago | (#35626762)

Yep - this is certainly a very impressive product. MIcrosoft have an absolutely world-class reseach lab/staff, but it seems rare so far for their work to make it into products (same as was the case with Xerox PARC).

Re:Very impressive (2)

Clsid (564627) | more than 3 years ago | (#35628332)

Yeah, they make nice products when they face competition, there is no doubt about it. But even then, some of the commercial practices are questionable and that's where most of the hate comes from. For instance, you buy an XBox360 and a PS3. In the XBox you have to pay a monthly fee to play online games where as in the PS3 is completely free. If Microsoft is the only player in town in that particular case then we would be in a world of hurt. Luckily, having options pushes Microsoft to do the right thing, even if it doesn't end up doing most of the time. To me, that's what Linux and free software is really all about, more about options instead of being a much superior product.

Re:Very impressive (1)

ObsessiveMathsFreak (773371) | more than 3 years ago | (#35630454)

The Penny Arcade strip was actually a send up of all the ridiculous hype surrounding the device. It can't actually restore sight to... you know what, never mind. Yeah, the Kinect is the digital manifestation of the second coming. It is the apex of technological development for the human race.

Really? (0)

Anonymous Coward | more than 3 years ago | (#35631916)

Way more hyperbole was used to describe the feats of the Wii remote than in the Penny Arcade comic, even before Motion Plus was released...

Summary hyperbole (1)

Anonymous Coward | more than 3 years ago | (#35625466)

I haven't thoroughly read the paper yet, but calling this an AI breakthrough is inappropriate for a number of reasons. First, this is an application of machine learning, which is not the same thing as AI. Second, it seems to be a fairly incremental work building on very common techniques--very far from a breakthrough in any respect. If you don't believe me, see some of Jamie Shotton's other work [shotton.org] , which is good work, but this is nothing extraordinary in comparison.

Re:Summary hyperbole (0)

Anonymous Coward | more than 3 years ago | (#35625768)

AI is a truly huge subject that has separated into many fields like natural language processing, computer vision, and machine learning. It definitely counts as a breakthrough in the sense that the techniques developed are finally reaching the consumer field. Consumer products are always going to be lagging behind cutting-edge research, so of course some researcher's homepage will have stuff way more recent than what the Kinect employs.

Re:Summary hyperbole (1)

Needlzor (1197267) | more than 3 years ago | (#35627714)

As Abe Othman and Ariel Procaccia said: "AI is whatever gets published at AAAI/IJCAI". Best definition of AI yet.

Re:Summary hyperbole (1)

Jeremi (14640) | more than 3 years ago | (#35626962)

First, this is an application of machine learning, which is not the same thing as AI.

That's the beauty and mystery of AI -- once a technique is actually made to work on computers in the real world, it loses its status as an "AI technique". The AI goalposts automatically move ahead to some other, harder problem that isn't solved yet. Eventually we will have HAL-9000 style computers everywhere, and people will continually piss them off by telling them the reasons they don't count as "real AI".

Re:Summary hyperbole (1)

Anonymous Coward | more than 3 years ago | (#35627076)

Nah. twice, three times, max. then the people will start dying.

Developed by a 3rd party? (-1)

Schuthrax (682718) | more than 3 years ago | (#35625824)

I thought everything was developed by a third party? So why is MS getting all the credit? Not to mention this doesn't sound like a breakthrough since it is using some common, and old, techniques.

Re:Developed by a 3rd party? (0)

Anonymous Coward | more than 3 years ago | (#35625864)

The need to process 200 frames a second kind of puts a limiter on what techniques you can use.

Re:Developed by a 3rd party? (2)

shriphani (1174497) | more than 3 years ago | (#35625974)

The sensor came from primasense. The algorithms in it are entirely from MSR.

Re:Developed by a 3rd party? (0)

citizenr (871508) | more than 3 years ago | (#35628030)

The sensor came from primasense. The algorithms in it are entirely from MSR.

Why can you download SDK with this software from Primasense directly, but not from Microsoft if the algorithms are M$ property?
Looks like M$ is just appropriating third party research.

Re:Developed by a 3rd party? (1)

shriphani (1174497) | more than 3 years ago | (#35628224)

Sorry I misspelled the name there. The company is PrimeSense. Here's where I see the paper beating the OpenNI SDK - 200 fps on consumer grade hardware. This is just what the paper claims it is - a simple machine learning technique that when applied correctly produced very good results and allowed them to launch a highly successful peripheral.

Looks like M$ is just appropriating third party research.

Splendid. Primesense are not complaining about this paper but you accuse MSR of stealing work?

Re:Developed by a 3rd party? (1)

marcansoft (727665) | more than 3 years ago | (#35628692)

PrimeSense developed the sensor technology (hardware and firmware) that gives you a depth image. Microsoft took that depth image and created the algorithms that perform body tracking (software).

PrimeSense also have their own body tracking solution (they call it NITE), but it's based on an entirely difference concept and requires a calibration pose to "lock in" initially. Microsoft doesn't use NITE.

Kinect's Perspective (1)

AnotherAnonymousUser (972204) | more than 3 years ago | (#35626402)

So...it can't see the forest for the limbs?

Summary in a few words (1)

elsJake (1129889) | more than 3 years ago | (#35626852)

Neural Network / perceptrons.

why forests (0)

Anonymous Coward | more than 3 years ago | (#35626960)

I would like to know why they choose a relatively slow method, random forests (RF), over something like an SVM or another classifier based on a convex optimization? They claim the RF run fast on the GPUs, although they also mention reference [6], which uses SVMs on this problem. Are the RF actually faster than a modern, non-linear SVM implementation?

Amazon Mechanical Turk involved? (0)

Anonymous Coward | more than 3 years ago | (#35628400)

And is this how they got the images tagged in the first place? http://kotaku.com/#!5605936/is-this-how-microsoft-will-fix-kinects-couch-problem

TFA makes it sound like they're cheating (2)

L4z4ru5 (1705054) | more than 3 years ago | (#35628454)

"[..] the decision trees were modified until they gave the correct classification for a particular body part across the test set of images"

this is called cheating in machine learning (you are not allowed to modify your model(s) based on the results on the test set).
and of course it is not what they do.

nice piece work, tho IMHO not AI breakthrough.

Re:TFA makes it sound like they're cheating (1)

flyingkillerrobots (1865630) | more than 3 years ago | (#35630620)

That's what a tuning set is for. Do you trust the summaries here to give a perfect description of what is going on?

Thanks for NOTHING, Microsoft! (0)

pandrijeczko (588093) | more than 3 years ago | (#35628972)

Being possessed of an enormously large penis, I am unable to use Kinect as it keeps detecting it as a third leg!

Re:Thanks for NOTHING, Microsoft! (0)

Anonymous Coward | more than 3 years ago | (#35629474)

Isnt that your walking stick gramps?

Re:Thanks for NOTHING, Microsoft! (1)

Neil Boekend (1854906) | more than 3 years ago | (#35665278)

There are 2 solutions:
  1. Don't play naked.
  2. Don't lie about your penis size

Operating system? (-1)

aled (228417) | more than 3 years ago | (#35629086)

That 1000 core system used to train Kinect... what version of Windows was running?

Re:Operating system? (-1)

Anonymous Coward | more than 3 years ago | (#35631222)

That 1000 core system used to train Kinect... what version of Windows was running?

It was probably running Linux :)

Decision tree my a$$ (1)

sundru (709023) | more than 3 years ago | (#35629114)

The method they are using s called as haar cascades postulated by viola jones. I have used the same with opencv for a bit now. http://en.wikipedia.org/wiki/Haar-like_features [wikipedia.org] It's basically passing An image thru progressive classifiers to get a final weight of match. Microsoft may have done the training for generating the classifiers but the method has been around for a bit. "Decision tree".... Pfffft.

Why? (0)

MoeDrippins (769977) | more than 3 years ago | (#35629346)

Why can't MS do stuff like this in all their departments? Are there not enough smart people to go around? You get truly cool things like this, juxtaposed with
lame "us too!" attempts like WP7 and Bing.

Re:Why? (0)

Anonymous Coward | more than 3 years ago | (#35632806)

From what all the reviews say, WP7's Metro UI is anything but a "me too" attempt. It's often touted as being the best UI out there (I suspect just because iOS and Android UIs are rather similar and anything different is refreshing). Also, there's only so many ways you can skin a search engine, so being "me too" is kinda expected. Yet I find Bing does do much better than Google at some very few things, e.g. travel search.

Also, most of MS's cash cows are office software... Not much scope for doing cool stuff there.

Re:Why? (1)

im_thatoneguy (819432) | more than 3 years ago | (#35634186)

"Us too!"?

Well it's hard for them to do stuff like this in all departments when you don't acknowledge all the other times that they offer innovative or superior products.

WP7 is in my opinion a far better thought out operating system from a user standpoint than any of the alternatives. So if by "Me Too!" you mean they released a great rewrite of their product which has been on the market longer than either Android or iOS then yes they too continued innovating. WinMo go sucky but when it was released it was pretty amazing. The problem was that A) capacitive touchscreens were prohibitively expensive so styluses were the only useful input device and B) Data Plans were prohibitively expensive and painfully slow.

They should have started preparing for the day when finger input would be useful and data plans would be accessible sooner but they eventually got caught up.

Zune would be another example where Microsoft was both releasing tech before the ipod and with the Zune still offering a superior product. The fact that it didn't sell well had far less to do with the fact that it was a bad product than it just didn't have the brand recognition when it launched as the ipod.

left or right? (0)

Anonymous Coward | more than 3 years ago | (#35630764)

So which side does it use, left or right [heraldsun.com.au] ?

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?
or Connect with...

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>