Slashdot Mirror
Seeing Around Corners With Dual Photography
An anonymous reader writes "This project (which is part of this year's SIGGRAPH) has absolutely blown my mind. Basically they photograph an object with the photosensor at one point, and the light projector at another, and use the Helmholtz reciprocity algorithm to virtually switch the locations of the camera and projector, showing exactly what the light source "sees"! If that doesn't make sense to you, check out the research page and make sure to watch the 60MB video at the bottom. The playing card trick will leave you speechless!"
Suppose you shine a projector upwards from the ground.... and take a photo of a girl... what will the technique generate?
Online backup with Mozy, sounds like Ozzie, but more!
Where does seeing around corners come in?
With a video projector providing structured illumination, reciprocity permits us to generate pictures from the viewpoint of the projector, even though no camera was present at that location.
Other than using electrons instead of light, that's how a scanning electron microscope works. An object is scanned (raster scan) and one or more sensors near the target pick up the reflections to generate an image. In the SEM the image appears as viewed from the scanning electron beam source.
In the optical one mentioned in the article, the light source is a raster scanning projector which lights a target. The image is produced from photodiodes picking up reflected light.
These two systems are very much alike. One uses photons and the other electrons. The end image is generated the same way.
The truth shall set you free!
Note: I haven't read the paper yet, but it is downloading.
It seems like this might have some military applications as a result. Imagine sticking a photo-resistor array under a door or through a window and then getting "viewpoints" from any of the lights in the room. Could aid in target aquisition and elimination.
Not sure how well it works for something like that, but this is a rather impressive (at least to me) research project.
There are only 10 kinds of people in this world... those who understand binary and those who don't
1. Reverse transformation for any interesting case (note that no places are actually revealed on their example!) will always be close to singular, that means in practice that your noises (due to raster, finite precision, and just measurement error) will eat any signal in result.
2. You should know not only amplitude, but *phase* of the source signal, that means for light that you have to use coherent light source and utilize interference on the receiver.
1 + 2 = holography, so what is new?
(Read the article, but still downloading the movie)
> extrapolate what someone is watching/reading/viewing on screen?
h tml
Something like this?
http://www.cl.cam.ac.uk/~mgk25/emsec/optical-faq.
Voltara
http://graphics.stanford.edu.nyud.net:8090/papers/ dual_photography/
Come on kids, coralcache is the way to go. no more direct linking to servers that go down quicker than, well, you know.
I'm wondering if there isn't some way to semi-automate the torrenting process.
.avi .mpg .qt etc to the torrent client.
I don't see any reason a torrent client can't be set up to allow a HTTP seed in addition to all the torrent peers and seeds. Granted it's going to get very poor speed, but as soon as a chunk makes it out into the swarm it should disperse to everyone fairly rapidly, and the more automated it is, the sooner there will be other seeds to take over.
You would still need a database somewhere to provide a URL to torrent mapping, but perhaps something like the new distributed DB in the most recent Azureus would be flexible enough to encompass the task.
Once you have those two pieces in place, it's as simple as reassociating browser links to
"We are what we pretend to be, so we must be careful about what we pretend to be." --Kurt Vonnegut
how about a dark room with a tv or CRT monitor on? could a simple light sensor (maybe the remote control sensor? or would it not work with IR?)allow a tv to function as a camera? hello george orwell!
Which gets me wondering: say you can see in someone's window, but the view is not very interesting: you only see a section of wall; everything else in the room is out of view. But: there is a CRT TV on in that room, and you can see its reflected light on the wall.
How much information can you gather from that reflected light?
You could of course recinstruct the image on the CRT, but that's not very interesting.
The TV does not scan a focused image on its surroundings like the projector does, so you couldn't get a TVs-eye view of the room witht eh same technique.
OTOH, it is clear that from sampling even just a single point on the wall, you could get a silhouette of anything occlusion over the screen seen from that point. At least provided you had a pure white image on the CRT, OR knew what image was on and could calibrate for it.
How far could you get with all the information escaping the window in your direction?
sudo ergo sum
Several projects at SIGGRAPH last year addressed the question of what you could do with a planar array of cameras. You could consider this the natural extrension of stereoscopy (two cameras) or a cost-effective approximation of real-time holography. Some of this research is motivated by that commodity digital cameras and real time digital image processing computers can be bought at low prices, and assembled like RAID disk arrays or cluster computers.
Applications of these arrays included several kinds of real-time 3D TV (without silly glasses). The Stanford group pushed "conformal imaging", that is a cube of image planes at various depths and all viewpoints. This has the effect of looking around corners and through keyholes: if there a path for light to get through, you can probably extract a complete image. This does involve some mathematical massaging of multiple-camera images. Cheap Graphical Processing Units (GPU) from game machines can be reprogrammed to process images in real-time.
I can't RTFA, but I'm pretty sure that what you describe is not what they're doing. The remarkable claim that they make is that from images of a three-dimensional scene that are captured at a particular camera location, they can render an image that the camera would have seen from a different location (namely the location of the illuminator). Furthermore, they do this without a priori knowledge of the scene geometry. In your barcode example, you need a priori knowledge of the position of the source and the camera to correctly re-render the image from the perspective of the source.
"It take 9 months to bear a child, no matter how many women you assign to the job."
They linked to Stanford.
/. Stanford. This is not Podunk U!
Who would imagine that we could
Oh well, I guess the Graphics department at Stanford isn't recieving any love from their IT department.
Need help treating your acne? Come here!
But there are two catches: (1) when you see a point in the scene it might not be along the ray you expect from the projector because it might be due to a reflection and (2) it's expensive to rasterize each point individually so instead you use a binary coding scheme to you only have to project log(N) images, where N is the number of pixels. Dealing with these issues successfully is very cool - but fundamentally the original poster is giving a 100% correct description. This talk of "Helmholtz Reciprocity" gives the misleading impression that something deeper is going on. "Helmholtz Reciprocity" is why ray-tracing works, ie. why you can send rays 'backwards' from the camera into the scene to generate a CG image, it's not some deep new principle.
Doesn't it make you feel good to know that our freedoms are protected by politicans, lawyers and journalists.
We didn't /. Stanford. Almost all the research groups in the CS department run their own servers and the same is true of the graphics folks. It's simply one server that's being hammered and it can't handle the capacity. Bandwidth and network latency are fine - just the server itself does not have enough processing power/memory to handle all the requests (it's probably not much better than your desktop).
By the way, one of the guys, Levoy, is awesome. He did all that digital modelling of the statue of David stuff.
Yep. This is why Slashdot should cache pages along with the associated images and videos. Presto! No more slashdot effect. And saying "But what about mirrordot?" is not valid, because people only go to mirrordot after the original sight has already been crushed into oblivion. And the argument in the FAQ is total BS, too. Oh sure, Slashdot is really concerned that the site in question won't get its precious ad revenue when people are viewing the cached version.
NEWS FLASH: The only people who will be viewing the cached version are... wait for it... Slashdot users! That's right! And more importantly, you don't generate a lot of advertisement revenue when your site is offline due to the Slashdot effect.
So the Slashdot editors' argument (here) basically boils down to this:
Private: Sir, we've accidentally launched a nuke that's headed for downtown Maimi.
General: Boy, that sucks for Miami.
Private: Well, sir, we've got twenty minutes before the detonation -- shouldn't we at least sound some sirens or something and at least give them a chance to evacuate?
General: Sure, I know that evacuation sounds like a great idea, but think about it -- you'd be depriving all those people of their right to see the beautiful mushroom cloud that forms. And anyway, lots of people will probably survive the explosion. Only the unfortunate (half million or so) people who live right in the downtown area and don't have proper nuclear-bomb-proof apartment buildings will actually die. I mean, hey, maybe we could try to just evacuate those unfortunate few, but do we really want to go to all that trouble? In the end, private, evacuating Miami is "a complicated issue that would need to be thought through in great detail before being implemented."
Private: Excellent point, sir. Poor bastards.