3D sterio pairs using shifts and swings?

[Drew Bedo]

The DIY forum has a thread on building or modifying a LF camera to produce images that are stereo pairs. The suggestions involve hardware such as sliding backs, Dual lens lens boards and so on . . .It is a DIY forum, right?

This got me thinking. Is it possible to create stereo images by using the movements of the camera? Front and rear shifts along with skight swings might also work—or would it?

I can think of reasons why this might be time consuming or fussy to set up, but will it work?

[Two23]

I started shooting panos with my Nikon 24m PC-E on Nikon D800E by taking three shots & lens movements. This works very well! I also thought about doing what you're talking about with my Chamonix 045n, but I think you would also need a septum to keep half the film sheet in the dark.


Kent in SD

[Graham Patterson]

You just need to place the optical axis of one frame separated by the right amount (based on eye separation and viewing distance) from the other. You might have enough shift front and back on a 5x4, depending on the camera. You would still have the problem of taking two exposures separated by time*, which is the biggest issue when using one camera for stereo pairs. Interiors would be a good place to start.

Try it 8-)

* Insert dark slide, reverse holder, pull dark slide, cock shutter, apply front and rear shift to opposite side without disturbing the other alignment, trip shutter. Should be possible with enough motivation.

[Drew Bedo]

Oh sure, using movements would be a little cumbersome. No question about that. I was just spit-balling some thoughts about how to do it without a hardware solution.

Perhaps the simplest hardware approach would be a lateral shift of the whole camera on a rail.

I have had only one eye working right most of my life (no binocular vision) so stereo viewing is just theory to me.

[BrianShaw]

...

Perhaps the simplest hardware approach would be a lateral shift of the whole camera on a rail.

...

That is the traditional solution when a two-lens camera isn't used. There is likely a reason why it is the traditional solution.

[BrianShaw]

Have you thought of the notion of taking a single image, then cropping into two that approximates being shot separately and laterally shifted?

[Jim Jones]

The stereo effect is derived from near subjects being shifted in relation to distant subjects. The only way to achieve this with a single lens in a fixed position would involve manipulating the image in a photo editor.

[DrTang]

get or make a slider bar - they are not too expensive or hard to make

but.. if you are going to take pix of anything moving..like people - - forgetaboutit

Polaroid made an ID or passport camera with two lenses and a septum for side by side shots on polaroid sheet film at one time..I was going to see if I could make that into a 4x5 stereo camera - alas, another project sold off after sitting around for a couple years

5x7 stereo cameras were relatively common at one point - graflex made one I think - maybe search around so as not to have to re-invent the wheel

also.. have you considered making red/green prints from the negs and having a bunch of glasses about for viewers? don't know if it would work...but maybe

[Drew Bedo]

Have you thought of the notion of taking a single image, then cropping into two that approximates being shot separately and laterally shifted?

Imagine looking at say, a small house with a front porch; the viewing position is center front. In a stereo pair of images, the image from the left will show the front door displaced a bit to the right and the porch steps angled a little to the right. The right side image will show the same displacements to the left. When these images are properly dislayed and viewed from the right distance with both eyes, the impression is a three dimensional view from a flat representation.

There is no cropping or computer manipulation of a single image that I know of that will replicate that.

Can someone chime in here with some theory or formulae to make me seem like I really know this stuff? Working with one eye pretty much blind and the other corrected to 20/200 here.

[Drew Bedo]

Perhaps a more distilled example would be looking at two poles (telephone, flag etc) one directly behind the other. With the right geometry of lens separation and camera to subject distance, a stereo pair of images may reveal the rear pole a bit on the left side of the front pole, while the other image may reveal it a little on the right.

Again: I welcome clarification.