Can we make a 3d project without 3d rig?

[Minu Park]

It seems, what 3d rig does is that, it captures the verity of depth of the field.

In that case, if I stop all the way up, and capture greater depth of the field (For instance using medium format F level 32), can't I make a 3d image when I divide two artificial focus range??

If my theory doesn't make any sense, I think I gotta research about them.

 

[Joseph Coleman]

Are you talking about a stereoscopic 3d rig? for 3d movies?

 

[Marc Wielage]

You can always shoot flat and then have one of the dimensionalizing companies like Legend 3D convert it...

http://legend3d.com

http://www.stereodllc.com/3d-conversion/

http://www.primefocusworld.com/3d-conversion

Many, many, many Hollywood blockbusters have gone this route, including The Avengers and Alice in Wonderland, both of which made over a billion dollars apiece.

Note that typical conversion prices range around $50,000-$150,000 per minute, depending on complexity. Facilities take about six months to convert two hours of material; if you try to do it faster, the results are compromised.

 

[jake blackstone]

3D camera system with a single camera already exists.
http://www.hollywoodreporter.com/behind-screen/nab-fraunhofer-previewing-hybrid-2d-435669

 

[Björn Benckert]

It's a bit depending on the imagery that you shoot, how complex it is to divide up in layers and so on. Basically everything else than two lens nodal points is the same as faking it and normally that comes with some artifacts. If you all ready shot the material it's easy to get a quote and or see test results. If you are about to shoot it I would recommend a stereo rig or at least second pass with a camera offset. We have done quite a few mono to stereo shot so if wanted send me a link and I can give you a quote for fixing it. Normal price range is about 2USD / frame, but again it's always depending on whats in the scene.

 

[CK Olsen]

You can't capture stereo 3D with a single viewpoint lens.
You can capture stereo 3D with a single camera, as long as you ALSO use a specialized lens splitter, or a twinned lens with a a single mount, but these are essentially variations on the 3D rig you are asking about.
Yes, you can create the feel of stereoscopic 3D in post from a single viewpoint image stream, but creating the look of stereo 3D is far different than capture.
Try capturing stereo audio with a single mono microphone sometime if you don't understand.
The stereo effect results from subtle differences in the placement of the capture apparatus... two eyes set slightly apart horizontally... two ears set slightly apart, aimed in opposite directions. Multi Lens 3D rigs work to recreate this multi-viewpoint capture apparatus.
-CK

 

[Keith Putnam]

It seems, what 3d rig does is that, it captures the verity of depth of the field.

A 3D rig captures parallax information; two different views of the same scene offset horizontally. This is what your eyes do. One eye, or camera, sees "around" one side of an object while the other eye, or camera, sees "around" the other side of the same object (to radically simplify the situation). A 3D rig allows the two cameras to move horizontally, which is called Inter-axial distance, and to "toe-in" toward one another, which is called convergence. Inter-axial distance is related to the amount of apparent spatial volume captured by the system while convergence defines the screen plane; that which is beyond the convergence point appears to be "behind" the screen while that which is in front of the convergence point appears to be "in front of" the screen. None of this is directly related to depth of field, which is how we describe the total amount of distance from the image plane which appears to be acceptably in focus.

As others have mentioned, if you want to create a stereoscopic image from a single source, you must either "dimensionalize" the 2D image by creating parallax information, usually by a combination of painstaking manual rotoscoping and automated rotoscoping, or you can capture a "depth map" of the scene and have animation and compositing software essentially use your 2D image as a texture on a digitally generated 3D environment.

 

[paulherrin]

it may be possible to simply shoot side by side with 2 cameras (instead of a mirror rig) and create a virtual 3d scene to "re-shoot" with your settings dialed in digitally... anybody tried it?

 

[Björn Benckert]

it may be possible to simply shoot side by side with 2 cameras (instead of a mirror rig) and create a virtual 3d scene to "re-shoot" with your settings dialed in digitally... anybody tried it?

yes very possible but you run into issues when things get to close to the two cameras. If you would put two epics right next to each other I would say you are safe as long as main interest in the scene is no closer than about 2-3 meters from cameras.
if you build such rig I would recomend to make it so both cameras shoot straight or have a comon interest point about 20 meteds from camera. And then you use slightly wider lenses than wanted and do the "toe in" / adjust the interest point in post. Normally we do this in nuke but just about any 3d compositing tool can do the adjustmen. I actually think you can do it in rcxp but i never tested myself.
one important thing is sync between the two cameras, if you only want to play around with it you can cheat sync with high framrate and then in post sync it as good as possible. A really cool 3d effect can be made by shooting big vistas and have a big distance between the two cameras, like 2meters or such then that makes the landscape look like a miniature as the stereo effect is basically scaled…

Edit: Learning by doing is a beautiful thing, lol: http://www.hdslr-cinema.com/news/technical/syndicate-dual-7ds-for-3d-stereoscopic-movies/

I been quite deep into Stereoscopic stuff, but I never really learned to like it, no matter how good it's done it always hurt my eyes and personally I'm happy that it seams to be a trend that will fade out.

 

[paulherrin]

interesting, if a bit off topic

http://www.fxguide.com/featured/disneys-new-production-renderer-hyperion-yes-disney/

Disney has a very long and proud history, especially in recent years with stereo production pipelines. One of the advanced techniques used in say a film such as Tangled, was to split the image planes depth wise for some character shots. Since a hero character close to camera felt attractive with a stereo convergence, that would mean the background in the same shot would be too extreme and unpleasing on the eye. Instead of dialing down the stereo effect overall, Disney pioneered the use of special camera rigs that would render the foreground with one ‘stereo camera’ or solution, while rendering the background with another. Thus the foreground character would appear round and with more fullness, but their background would appear more relaxed, less stereo pronounced. So successful were these stereo camera rigs that Disney Animation provided them to Pixar, and they were adopted in Pixar features also.

basically what i'm thinking is that capturing side by side (enough to get 'around' your foreground, to assist or even circumvent extensive paint/roto) and of course augmenting that with depth map and/or projection (and stereo compositing) techniques... and this idea from disney is a very good one to go with as well in the virtual world. i've heard about proprietary programs that can create rich 3D (not stereo) from multiple synced cameras, beyond basic point clouds... when this advances that will probably be the way to go, i.e. red's mantra of processing in post. and then we'll finally be back towards directing plays and worrying less about fiddling with distracting stuff on set.

 

[Marc Wielage]

3D camera system with a single camera already exists.

Naaa, just a proof of concept device. I agree that it'd be an interesting way to shoot: one hero camera, and then little tiny cameras anchored nearby that just capture depth information for the dimensionalizing people. It would be a far, far more accurate way to fake 3D than the VFX methods used now.

I been quite deep into Stereoscopic stuff, but I never really learned to like it, no matter how good it's done it always hurt my eyes and personally I'm happy that it seams to be a trend that will fade out.

There are still some major films coming out in 3D and being produced in 3D now. The last time I checked, I think there's at least 11 $200M+ live-action 3D films coming out next year. The thing flogging 3D is exhibitors who want to be able to charge a little more money for tickets. Interestingly, 3D is still very big overseas, particularly in China, which is a huge market.

 

[paulherrin]

speaking of depth maps and info... am i remembering correctly that there is some development at red of motion z-channels?

that would be another good starting point for dimensionalization. i think microsoft bought the ip of whatever company had been working on this for years...

btw, is anybody getting motion data from the 3-axis sensor?