Project Natal: MS Full Body 3D Motion Detection

[liolio]

Actually proper order recognition would be great:
In ME for example
Wrex attack (attack the character you're pointing at)
Liara singularity
pause
lock
Fire
etc.
A pretty neat way to navigate though other wise pretty complicated menu especcialy when you have a squad at your command.

 

[TheWretched]

Actually proper order recognition would be great:
In ME for example
Wrex attack (attack the character you're pointing at)
Liara singularity
but it
pause
lock
Fire
etc.
A pretty neat way to navigate though other wise pretty complicated menu especcialy when you have a squad at your command.

Endwar does it like this... only tried the demo once though... and I had problems using the voice commands some times, as it didn't want to recognize some commands.

Thing is, the Milo demo showcased natural language recognition and Molyneaux said (iirc), that it also recognizes emotions... I call bullshit on both fronts here. It "might" pick up anger or happiness, but the other emotions like fear or sadness aren't as clear cut.

And Microsoft has already shown us how well their voice recognition works in a less perfect environment with their Vista presentation 2 years ago.

 

[V3]

Any optical solution Sony can use, MS can use too. Unless Sony's implementation includes some other tech like a transmitter in the target device, they have no advantage.

Well as far as I can see, the PS Eye should be capable of doing what Natal do too. The Sony solution is essentially Natal + Wiimote.

I was just curious about how much more accurate Natal is. After reading from the blog, it is not as impressive as I first thought.

Hands free gaming was made popular by Eyetoy. It's just Sony doesn't have the development house of Nintendo or even MS. The games suck. In the end it's all in the games. Wiimote wouldn't have been a success without Wii Sport.

Both Sony and MS were just showing technology. Nintendo actually showed a console saler game, when they demoed Wii for the first time.

 

[arhra]

Thing is, the Milo demo showcased natural language recognition and Molyneaux said (iirc), that it also recognizes emotions... I call bullshit on both fronts here. It "might" pick up anger or happiness, but the other emotions like fear or sadness aren't as clear cut.

Read through the Eurogamer interview, he's remarkably frank about exactly what it's actually doing and how much of it's just a convincing fake:

Eurogamer: So what can Milo do?

Peter Molyneux: Milo can recognise the emotions on your face and the emotions in your voice. He can recognise certain words you say. You can have conversations with him, you can read stories to him. We're trying to bring all these things together. Some of them are tricks - I'll be absolutely honest with you - to make you believe Milo's real.

He can recognise what you're wearing. If he notices you've got dark bags under your eyes he will say, 'You look tired today.'

[...]

Eurogamer: You said he only understands certain words. So presumably you can't have a conversation about the situation in Palestine?

Peter Molyneux: The number of words he understands is built up over time. For Claire [the lady who demoed a conversation with Milo during Microsoft's conference], it's something like 500 words.

But we haven't cracked the real problem, which is him understanding the meaning of it all. He'll give you the illusion he does that. The interesting thing is you can only talk to him when the Talk icon appears at the bottom of the screen. That's when he's listening to you; the rest of the time, he's not. He's listening to you because there's a context in which you can talk to him.

One of the journalists who came in before you had obviously read up on the Turing test. He asked Milo one of the questions in the test - 'Do you remember when we met yesterday?' Well, of course, we haven't cracked the Turing test. If we had, then applying it to a computer game would be the last of the solutions we'd use it for.

So they're not actually claiming anything that far out there; mostly just putting together a bunch of existing tech (a couple of quick google searches dig up at least one paper on facial emotion recognition, and a whole bunch on vocal emotion recognition) and using a lot of clever trickery to paper over the gaps and make it believable.

 

[RobertR1]

Read through the Eurogamer interview, he's remarkably frank about exactly what it's actually doing and how much of it's just a convincing fake:

People need to read and interview over and over and OVER. The amount of dialouge would be cut back severly as most of it's silenced by what he describes in detail. MS and Peter are not trying to take people for a ride. They're not running around screaming "WE ACHEIEVED AI and it is perfect during interaction!" What they're showing you is interaction progression by a new method, how they plan to approach it and what they're hoping to bring to the market. By Fall Christmas 2010, no doubt, they'll make good progress and have a working product for the customers. I'm always amazed that the biggest software company in the world with a r&d group larger than most companies get discounted so often, esp on the software side! While the xbox is a good consumer showcase for such tech, I have no doubt other areas of the CE world are closely keeping an eye on this. This tech is beneficial to MS overall, not just in the gaming division.

 

[Asher]

Well as far as I can see, the PS Eye should be capable of doing what Natal do too. The Sony solution is essentially Natal + Wiimote.

This is not at all true, and it's been discussed to death.

Natal is not simply a webcam.

 

[RobertR1]

This is not at all true, and it's been discussed to death.

Natal is not simply a webcam.

Until we have the end product close to ready for launch, for many it'll be what they want it to be, for the sake of their mindset. Some will think the world of it, while others will try to downplay it. Natal is going to quickly (alread has been) going to turn into a team sport where you pick a side and bang heads till retardation.

I'm just happy to see the positive impressions being posted by various sites. That's a good start. We'll see where it ends.

 

[obonicus]

People need to read and interview over and over and OVER. The amount of dialouge would be cut back severly as most of it's silenced by what he describes in detail. MS and Peter are not trying to take people for a ride. They're not running around screaming "WE ACHEIEVED AI and it is perfect during interaction!" What they're showing you is interaction progression by a new method, how they plan to approach it and what they're hoping to bring to the market. By Fall Christmas 2010, no doubt, they'll make good progress and have a working product for the customers. I'm always amazed that the biggest software company in the world with a r&d group larger than most companies get discounted so often, esp on the software side! While the xbox is a good consumer showcase for such tech, I have no doubt other areas of the CE world are closely keeping an eye on this. This tech is beneficial to MS overall, not just in the gaming division.

How about we let them deliver the working product first, since we're talking about Molyneux here. For the record, I don't doubt Natal will work as well as the dodgeball demo, or even better.

And hey, if we can keep expectations in check and not start making plans for all the free time they'll have when Natal delivers their robot butler, maybe the naysayers wouldn't have to do so much naysaying.

 

[Asher]

For me, I don't really care for anything in Peter's presentation. I recognized it as smoke and mirrors from the beginning, probably because I had some semblance of how it all works.

But the potential is still there and enormous, especially with the library of MS Research technology MS is sitting on related to sound and video processing, and their track record of SDK development.

We're not going to see a fully conversational AI within the next decade, at the very least. People need to accept that and move on.

What we will probably see are keyword recognition and its integration into dialog-tree like conversation a la Oblivion and Mass Effect -- in terms of audio. Add in combination of things like using the camera + sound positioning to have the avatars have a good idea where the user is standing/sitting, and having the avatar's head track the user would go a long way to adding some immersion to these games. And that's just for games that'll still use the classic 360 controller too. I still need time to digest what'll be possible for games using other control mechanisms. That's a blank canvas with Natal.

 

[RobertR1]

How about we let them deliver the working product first, since we're talking about Molyneux here. For the record, I don't doubt Natal will work as well as the dodgeball demo, or even better.

What did I say different? Read the post below where I specifically state the importance of letting them deliver the product (getting close to) and see what comes of it.

And hey, if we can keep expectations in check and not start making plans for all the free time they'll have when Natal delivers their robot butler, maybe the naysayers wouldn't have to do so much naysaying.

Sadly there's two camps on this side also. One side who foolishly believe they'll deliver such capabilities and the other side who have an agenda, take this stance and then act like the product is a failure because it didn't meet their expectation.

 

[RobertR1]

For me a combination of controller + Natal would be absolutely fantastic for hardcore games. However, for casuals and even certain game types for all users, I'd love to see no controller, Wii Sports, sports and activities like games.

A few years ago there was a boxing game in the arcades, still is one at Dave and Busters in Fremont, CA that did motion capturing. You stood under the sensors, put on gloves and boxed against the AI. It worked quite well and definitely well ahead of anything Wii boxing can do. An evolution of a game like that would be awesome using Natal. Those are my two expectations from this tech and from what I can see they both seem possible.

Here is the old arcade game I'm talking about: http://freemotionfiles.blogspot.com/2009/03/mocap-boxing-video.html

 

[obonicus]

What did I say different? Read the post below where I specifically state the importance of letting them deliver the product (getting close to) and see what comes of it.

No, what you're saying is to give them the benefit of the doubt. No, sorry. Peter Molyneux does not get the benefit of the doubt.

Sadly there's two camps on this side also. One side who foolishly believe they'll deliver such capabilities and the other side who have an agenda, take this stance and then act like the product is a failure because it didn't meet their expectation.

Agendas run both ways. We have people cheerleading motion controls who otherwise wouldn't care for them (and you're actually one of the few people who has been consistent on this issue, for what it's worth). We have people railing on about how voice controls are the wave of the future when at the same time I didn't catch much interest on this board for Endwar when it was coming up last year, even on a technical level.

 

[Asher]

I still do not comprehend the obsession with Molyneaux. There was a lot of content presented aside from his usual hyperbole. Milo will never be a shipping product, I think we all know that. Time spent arguing about it is time wasted.

 

[obonicus]

But the potential is still there and enormous, especially with the library of MS Research technology MS is sitting on related to sound and video processing, and their track record of SDK development.

I have my doubts that there's such a breadth of tech there, to be honest, but sure, we're both speaking from a position of ignorance so I'll believe you.

What we will probably see are keyword recognition and its integration into dialog-tree like conversation a la Oblivion and Mass Effect -- in terms of audio. Add in combination of things like using the camera + sound positioning to have the avatars have a good idea where the user is standing/sitting, and having the avatar's head track the user would go a long way to adding some immersion to these games. And that's just for games that'll still use the classic 360 controller too. I still need time to digest what'll be possible for games using other control mechanisms. That's a blank canvas with Natal.

Sure, I buy that we could do it. But it's sort of the same thing as for motion controls for me. Why would I want to do that? Is this something I can't accomplish with my thumbs? Why do I even need Natal for this? The 360 ships with a wired, personal headset -- much smaller chance of someone saying something that is picked up erroneously by the mic array. Why are we applauding all these gimmicky ideas? Every single time a game has a goddamned sixaxis 'pull the lever yourself' puzzle it gets lambasted, and rightly so -- how about we avoid the double-standard, then?

 

[obonicus]

I still do not comprehend the obsession with Molyneaux. There was a lot of content presented aside from his usual hyperbole. Milo will never be a shipping product, I think we all know that. Time spent arguing about it is time wasted.

Partly because not everyone agrees on it. Partly because Milo is what is being used as a justification for Natal to baffled journalists who ask the very sane question of 'what exactly are the game applications for this besides Flail Your Arms at Dodgeballs?' Hell, I do believe I'm having a conversation with someone in this very thread who doesn't share your opinion that Milo will never be a deliverable.

 

[Asher]

Sure, I buy that we could do it. But it's sort of the same thing as for motion controls for me. Why would I want to do that? Is this something I can't accomplish with my thumbs?

Simple -- immersion and intuitiveness.

Sixaxis was awful because it doesn't address immersion. It's just a twitchy tilt controller that has very, very few legit uses.

 

[obonicus]

Simple -- immersion and intuitiveness.

Sixaxis was awful because it doesn't address immersion. It's just a twitchy tilt controller that has very, very few legit uses.

I'd argue a fundamental flaw rather than one of implementation -- I particularly don't crave pantomime to achieve greater immersion.

 

[MfA]

Until we have full body haptics and force feedback I don't particularly see the need for full body control except in party games either ... but in the end the interface gimmick was seen as the main cause of success for the Wii (which I think isn't true, it was one part price and one part random fad) and they all want a piece of that pie.

It's not targeted at "us" gamers.

 

[phat]

Hmm... With a nanosecond shutter, 3dv can only resolve about 15 cm of depth, since light travels about 30cm per nanosecond (divide this by two because of round-trip distance to get depth resolution).

 

[Scott_Arm]

No, what you're saying is to give them the benefit of the doubt. No, sorry. Peter Molyneux does not get the benefit of the doubt.

Agendas run both ways. We have people cheerleading motion controls who otherwise wouldn't care for them (and you're actually one of the few people who has been consistent on this issue, for what it's worth). We have people railing on about how voice controls are the wave of the future when at the same time I didn't catch much interest on this board for Endwar when it was coming up last year, even on a technical level.

The Milo demo that was shown during the presentation was BS for sure, but the one on ones the journalists are getting seem to be fairly honest and hopefully that means we'll get a good product in the end.