Slashdot Mirror
Disney Algorithm Builds High-Res 3D Models From Ordinary Photos
Zothecula writes "Disney Research has developed an algorithm which can generate 3D computer models from 2D images in great detail, sufficient, it says, to meet the needs of video game and film makers. The technology requires multiple images to capture the scene from a variety of vantage points. The 3D model is somewhat limited in that it is only coherent within the field of view encompassed by the original images. It does not appear to fill in data"
Using the new technology added short circuit from Wally, and goofy rendered flawlessly into the movie preview
http://saveie6.com/
This is great for scenery, it is amazing how much effort goes into the background scenery that no one will really pay attention to, but if you get it wrong everyone pays attention suddenly.
You just copy and pasted right before the punctuation mark for the summary?
The 3D model is somewhat limited in that it is only coherent within the field of view encompassed by the original images. It does not appear to fill in data
Just have the CSI boys zoom and enhance. C'mon guys, they've been doing this for years.
The technology requires multiple images to capture the scene from a variety of vantage points.
So simply the big difference here seems to be synthesis of properly spaced stereo cameras by using cameras positioned anywhere.
But it sounds less impressive than some of the stereo movies released from older 2D movies where there wasn't any additional
cameras, and someone simply assigned depths to parts of the image and put CGI to work on it.
Sig Battery depleted. Reverting to safe mode.
The technology requires multiple images to capture the scene from a variety of vantage points.
That's cheating.
Disney's version is for making even cheaper knock-off game content.
See this page; the Campanile movie is from SIGGRAPH 97. How is Disney's tech different?
Yes, its fairly easy to build a 3d model given enough input and the right algorithms. Look at all the 3d scanner software that uses kinect and multiple frames to construct a higher resolution model.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
Now, if it filled in data, that would be worthwhile.
All we got here is some pretty standard stuff
It's probably not new.
It's automated, as it doesn't require someone to create a 3D model of the scene, which was done for that movie. There's more differences, if you'd bother to read about both.
...where construction crews are responsible for building our universe every second. Somehow our hero accidentally falls into that realm, and sees some whitespace or voids. One of the crewmembers tells him, when nobody goes there, they don't really have to build it. That when one of their crew is negligent, and forgets to put stuff where they're supposed to be; this explains those incidents when you could have sworn you placed that hammer on that table a second ago...
WARNING: Smartphones have side effects--most of them undocumented.
The name of the algorithm is called 'affine reconstruction' and is a fairly well studied algorithm in computer vision. It is great that Disney and co. are releasing software to semi-automate the data input and reconstruction.
http://www.123dapp.com/catch
Autodesk has a service already available that does what the Disney does, it's called Recap.
http://usa.autodesk.com/adsk/servlet/pc/index?id=21350337&siteID=123112
They have a cloud service that can make full 3D models from photos.
"Densely sampled light fields in the order of 10^9 light rays allow us to capture the real world in unparalleled detail"
So the article abstract claims their input capture is not exactly "Ordinary Photos".
I have a program from the mid 90's that I got from a book about VRML http://www.amazon.com/Teach-Yourself-Vrml-Days-Sams/dp/1575211939 which would turn say buildings in photos into 3d objects. I think it was only a demo so never really tired it out to see if it worked.
by TheSpoom (715771) Uncaring Linux user here. I have nothing to add to this but please continue. *munches popcorn*
This was on here months ago, and it wasn't even that great then.
is hi res 3d naked models of miley cyrus and selena gomez. woot woot!
Hey gramps, what the hell is a "Polaroid"? r * sin (Hemorrhoid) ?
Next you'll go blathering on about irrational things like "phone books", when everybody all knows they're called Kindles.
the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff
http://www.robots.ox.ac.uk/~gk/PTAM/
did it 5 years ago
Who logs in to gdm? Not I, said the duck.
I weep for this version of /.
It seems that all true geeks are gone, or if not gone they are now silenced.
Last night I made a comment about Snowcrash in the Mars Virus Crap Story (found it on idle) and nobody commented, not even the trolls.
I guess /. is too, well, (uggh, I hate to say it) mainstream now.....
Just like Bladerunner and everything else 80's.
Scheisse.
I wonder if all those frames that stored say, John Wayne, could be used to create a fairly good 3D likeness. If not now, maybe soon. Also, who would own the rights to those performances?
Gee, it would be great if they actually posted some 3D examples in their gallery. Instead we just get depth maps and reconstructions from the same POV as the original camera. I want to see the 3D models this thing generates!
If you weren't a dick you'd realize several others have been automating this idea for years.
See this page; the Campanile movie is from SIGGRAPH 97. How is Disney's tech different?
I saw similar technology at CMU in around that same timeframe (late 90s).
My memory will be obviously hazy here, but the resulting output was much less refined. A simple box-shaped house, for example, ended up having wickedly jagged walls. The technology showed promise, but it was far from realistic.
The Disney folks, while not inventing the tech itself, seem to have taken it a step further. Their key claim -- "Unlike other systems, the algorithm calculates depth for every pixel, proving most effective at the edges of objects" -- certainly jives with my memory.
I for one can't wait to buy Star Wars The Original Trilogy 3D on BlueRAY!!!!!!!!!!!!!!!!!! And if the box the new Trilogy with the old that's three threes. How could any not buy it.
...use 4D light fields...
Did anyone else notice flashing images on the screen while playing the video?
I can't get it to stop exactly on the image but it looks like some sort of white and red striped thing near 7 seconds in to the video.
Odd.
Wasn't this the first time? Photosynth demo.. http://www.youtube.com/watch?v=M-8k8GEGZPM
Overlap the photos you're taking by 60% & look at them through a Stereoscope... you get 3D.
http://en.wikipedia.org/wiki/Stereoscopy
I spent years making a film all from 3d scans of photos... you think I could ever get any mention on the front page of any good website, no way... vimeo.com/69136384 .... Disney does the same thing & doesn't even show it from other angles ++ they had a post a few months back like this. Perhaps I should try making a film for them?
One problem with cheap and homebrew CNC machines is that the cutting head loses track of its own position. The feedback from CNC machines comes to the computer from sensors on the position screws of the various axes. These can lose their calibration. I keep thinking that with this kind of technology, a computer can "visually" determine the precise position of the cutting head and also of the material being cut. It can update the computer constantly on the exact shape of the remaining material, and then compare it to the design and calculate what else needs to be cut away. Basically, all of the feedback can potentially be just optical, but of course, the computer can "eyeball it" much more accurately than a person. Because good cameras are cheap and CPU cycles are cheap, I think this could be the basis of a very flexible and affordable CNC machine. I suppose it would be important to have a good sawdust removal system, because that could potentially obscure the view - although I guess the cutting head itself could be used to sweep it away.
http://tech.slashdot.org/story/13/06/23/168218/disney-research-creates-megastereo---panoramas-with-depth
The basic technique for this was demonstrated at Siggraph some years ago, I don't remember which off hand but before 2003.
From Wikipedia:
Released by MetaCreations Corp. in 1999, this application allowed users to create 3D models based on one or more photographs taken from various angles.
Great program. Never understood why it died. Assume it was corporate hijinks.
Yay me!
In another news, the Sun is shining. I mean seriously, light-field based 3D reconstruction has been around for many years. Hell, even one of my colleagues has built a rotating table-based camera setup to capture images and create a full 3D model. Just google light fields 3D reconstruction or structure from motion and smell the coffee.
Yeah, great news.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
So art imitating life, or life imitating art? Just saw the Futurama episode Forty percent leadbelly in which Bender takes a 2-D photo of a guitar, and gets it duplicated by a 3-D printer. Seems we are in the future already!
/ The Arrow
"How lovely you are. So lovely in my straightjacket..." - Nny
You got some Reddit love awhile back so quit your bitching :)
I'm not sure exactly what Disney's developments are, but I just wrote this up: Open Source Photogrammetry Workflow