Optical Camouflage Puts Kinect Into Stealth Mode 60
UgLyPuNk writes "Takayuki Fukatsu, a Japanese coder who works under the name Art & Mobile, has done a bit of trickery with Kinect and openFrameworks. The peripheral will still track your movement and position, but turns your image nearly transparent. Take a look (it's particularly obvious at about 1:30):"
The background doesn't change (Score:5, Informative)
Re:The background doesn't change (Score:5, Informative)
It's a shade more sophisticated than that. I think he's using the existing make-a-texture-mapped-model-of-the-space code, but telling it to texture map anything that's non-background with an pre-existing image of the background. It's a cute project, obviously intended to recreate the look of the sci-fi cloaking effect, rather than do anything clever. After all, you could achieve a much more effective result by just replacing the feed with one in which the person was not present.
Re: (Score:2)
Well, clever's the wrong word. Functional. It is certainly clever.
Re: (Score:1)
Re: (Score:3, Interesting)
Re: (Score:2, Informative)
The pink border is the wallpaper of his desktop computer. He is taking a video to his monitor.
Re: (Score:2)
Re:The background doesn't change (Score:4, Interesting)
"pre existing image of the background" this makes it an epic fail, but cute project.
I'l be impressed if it generates the background without ANY reference images or reference data.
Re: (Score:2)
"pre existing image of the background" this makes it an epic fail, but cute project.
I'l be impressed if it generates the background without ANY reference images or reference data.
Except for the fact that he moves the camera, thus changing the POV. Whatever he is doing to achieve this effect is live.
Re: (Score:1)
The camera he moves is the camera taking the image of the screen. He is not using screen capeturing technology to post the image to youtupe, he is using an external camera to take screenshots.
Somewhere in the video he is moving to the back, and then the effect disappears completely.If he was using a live stream for the background then the fact that depth information was lost, he should have been appeared, From this fact you can assume his using a static image where he uses the depth camera to make a preada
Re: (Score:2)
Re:The background doesn't change (Score:5, Insightful)
Re: (Score:2)
Can't speak for GP but I've done better using just a JPEG of a background. You can't see when I move at all.
Re: (Score:2)
Can't speak for GP but I've done better using just a JPEG of a background. You can't see when I move at all.
Are you joking? The whole point is to be able to see when he moves. It's a special effect to show a sci-fi kind of "cloaking". Sure you could implement something similar with a standard webcam, but the novelty here is that he seems to use the Kinect's depth information to work out how much distortion/lensing effect to apply. Hence when he stands against the bookshelf in the background, he disappears completely.
Re: (Score:1)
Re: (Score:3)
But he does have a point. How is this using kinect's abilities? How is this different than EffectTV's PredatorTV filter? Is the guy just using all the late kinect dev rage to get credit for doing something simple and available for years, or he is really doing something new?
I can't really see how the kinect depth info could help. Now, the other guy with 2 kinect's providing a 3d view really did have something, and he could easily add a predator filter to make something invisible, as he would be able to see b
Re: (Score:3)
The depth info tells you what's non-background in the scene unambiguously, and allows the texture map to distort to follow the object, which is a bit closer to the way the effect was depicted in GitS and the latter MGS games.
Re:The background doesn't change (Score:5, Interesting)
Notice how when he walks far back, the distortion goes away and he disappears completely. What does that tell you?
So, he is using depth info, but what he is doing with it is rather lame. He still has a static image of the empty room, otherwise when the person went far enough back, he would have to appear uncloaked not disappear completely. Of course you would need 2 kinects and much more work to avoid the need for a static background image and just apply the "cloak" to objects nearer than the background. But that would certainly be cool.
What we have here, you can do better without a kinect by simple diffing of the background image. If he at least used the depth info to alter the distortion it would be interesting, but it seems to me that when he is walking towards or away from the camera the distortion does not change at all.
Re: (Score:2)
I saw that of course. But the distortion is either movement or color (in any case without an explanation of what he is trying to do we can't judge), because when he walks slowly forward or backward there "distortion" is simply a static offset of the background. With a simple camera you can have different distortions with movement or brightness or color etc, with a kinect you could make the body look like a lens and that is not done here.
Re: (Score:2)
Hmm, upon watching more of the video, he seems to stop the effect or "disappear" completely if he goes far back enough, so I guess that is a way he is using the kinect depth data. However, when he walks forward or backward, nothing changes on how the background is seen through him, I would have expected something like the effect of a lens moving closer or farther, in general different distortion depending on how close he is to the cam.
Anyway, without knowing what he is trying to do, I can't be impressed by
Re: (Score:3)
There has been a plugin for after effects to do this EXACT effect but in more detail and clairty for years. all you need is a green screen shot of the target and the footage you want the effect on.
Re: (Score:2)
Looks like it takes a static shot, and then projects it on top of the 3D data from the kinetic.
Re: (Score:1)
It's kind of a neat effect, but it's not apparent at all what the Kinect's tracking stuff is used for.
Re: (Score:1)
Re: (Score:1)
This looks somehow familiar... (Score:1)
I would say it's similar to the recent hack that uses TWO kinect devices working together, but this time the programmer has used to to simulate a "predator-like" effect. I might be wrong about it, but if you watch carefully there's an obvious alignment/sync problem inside the "predator" shadow with the actual background (possible due to the image coming from a different angle).
Cheers!
Re: (Score:2)
a bit slashdotted (Score:4, Informative)
http://www.engadget.com/2010/12/02/kinect-now-offers-a-stealth-mode-courtesy-of-optical-camouflage/
http://www.youtube.com/watch?v=4qhXQ_1CQjg
Re: (Score:2)
I can't watch this video (Score:3)
It's camouflaged by work's firewall :P
Griffin? (Score:1)
direct link to video (Score:1)
http://www.youtube.com/watch?v=dPV2jmgecfU [youtube.com]
IMDB Reference (Score:2)
Re: (Score:2)
Re: (Score:2)
I think he has a camera pointing in one direction and the Kinect pointing to another, so he is actually projecting the Kinect volume info over an arbitrary image with a cool effect, he is not really cloaking against his background.
Not much of a difference with current blue/green screen effects. Final result is exactly the same.
why do you need a kinect for this? (Score:1)
effectv has this effect for years. http://effectv.sourceforge.net/predator.html
no specific kinect feature needed btw
YouTube direct link (Score:2)
http://www.youtube.com/watch?v=4qhXQ_1CQjg [youtube.com]
Site (Score:1)
looks like their site has also been well camouflaged.
Simple Static Image (Score:1)
Really don't see *any* use to this what-so-ever. The only difference between it and a live feed is that anything "not" covered up is live (as long as it doe
Once you are invisible and use no controls... (Score:2)
Rule 34 (Score:2)
Re: (Score:1)
You lie alone in your couch, doing humping moves, while looking at the TV where an animated chick tells you that it feels good?
You jerk off, while an animated character undresses and tells you how big you are?
Seems sad to me, even by slashdot standards.
Re: (Score:2)
And still no Skype support (Score:2)
Why is this a slashdot story?? (Score:1)
This is easily done with a regular webcam (apart from the extra depth-data). This story is lame, even if it were on a M$ fanboy game-blog...