Stereo photography?

[steven_e007]

Hi,

I've trying to find anyone on the board who has experience of taking stereo images...not sure of the correct area to discuss it, really.

The story is this:

There is a massive craze in 3D films at the cinema at the moment.
Trouble is, I can't see 3D

I had a squint when I was little. I had surgery when I was about six to correct it and my eyes work pretty normally, now. I can superimpose two images, but I never developed the brain function to construct 3D.

If I look at those test charts of coloured dots with the green and red glasses on? Where you are suppossed to see the Bunny? Nothing. Same with side by side spectoscopes (the victorian way of doing it) or those jumbled up dot pictures. Absolutely nothing...

Except... I can in black and white. I was in an opticians about 2 years ago and instead of using the usual colour chart they used a black and white image of a fly and glasses with polaroid lenses. Suddenly WOW! 3D!
For forty years I thought I had no 3D perception and then suddenly - 3D! Difficult to decribe the shock I got from that...

No idea why B/W - polaroid works, but colour with Red/Green doesn't work at all.

Not sure what would happen with colour using polaroid filters or some other technique, either.

So -I want to experiment with my own eyesite a bit. I'm interested in knowing what I can and can't see and why. I'd really like to have a go at making some stereo polaroid images.

Anyone got any ideas how this is done? Anyone ever tried it? I've had a google, but couldn't find much. There's a fair write up of the theory in wikipedia, but nothing on DIY pictures (Plenty on anaglyphs, the red/green technique - very little on other techniques).

 

[phaedrus]

stereo vision

Mind you, I don't want to discourage you from your experiments, but from your very well described symptoms one can deduce that you have limited stereo vision.
The Titmus "Fly" stereotest that you saw threedimensional presents the two images at a stereodiscrepancy of 1000 seconds of arc. That's a lot. Any eye specialist (I'm one) can examine your stereo vision with tests of diminishing stereodiscrepancy to find out your treshold. The result is given in seconds of arc for the last test you saw thredimensional and is relevant for some occupational tests and aeronautical licenses.

 

[steven_e007]

Hi Phaedrus,

Thanks for the reply. I have looked at colour anaglypt tests where the subject have different discrepencies and I can't see any of them at all. The "fly" I could see very easily. I am assuming that I have seen a colour image of the same angle of discrepency so it appears that for some reason I can see Black and White better than colour. I wondered if this is because the rods in your retina are distributely more widely than the cones? Don't know.

I'd love to have a test on a synoptophor but alas, in the UK opticians are private practices primarily concerned with making spectacles and screening for problems. As soon as they find a convergence problem it would be off to see an optometrist (don't know if this is the same wherever you come from?) under the control of the National Health Service. Since I am only interested in my eyesite and do not have a real problem I couldn't expect to, and wouldn't be, refered to an optometrist to be tested. I could probably pay for a private consultation, but I'm sure that would be serious money - and I'm not that interested!

Anyway, my OP was to see if anyone has any experience of making BW polarioid stereogrammes. The internet is full of instructions on colour anaglypts from digital images using photoshop and a digital printer - which is not only not what I'm after but strictly off topic for APUG.

I was hoping to do things properly with some polaroid filters on my camera and under my enlarger

 

[filmamigo]

Can I ask... are you colour blind? That would impact your results with the colour 3D tests.

As for making stereographs, I love to make them.

For the camera, I have used an old 127-format stereo camera. I can still get 127 film locally, you can order it too. I scan the film myself. If you used a stereo Holga or a 35mm Stereo Realist, you could likely have the lab scan them for you. I have also used a pair of digital SLRs with matching 28mm lenses, which avoids having to scan the film.

To convert the individual images into a proper "Holmes" card (the Victorian stereo views format that I prefer) I use computer software. It's called StereoPhoto Maker, and it's free. With the software, I can generate high resolution .JPG of the finished Holmes card. I take it to the local photolab and have it printed out on 5x7, trim the top and bottom a bit, and have a very professional looking result. They work great in the classic stereograph viewers. Of course, black and white works just as well as colour, because the process is digital.

 

[filmamigo]

I have also considered using a Fuji passport camera -- the kind that takes FP100 instant film. They are by default a stereo camera, as they take two images side-by-side with two lenses. The drawback is the focal length -- they are intended for portraits, not scenics. I'm sure the resulting instant print could be split, trimmed and mounted on 4x7 card stock for use in a Holmes viewer.

 

[filmamigo]

One final thought... you could probably make a nice optical stereo print by shooting 35mm in a Stereo Realist, and enlarging it on a medium-format (6x7 or 6x9) enlarger.

 

[maciekz]

I'd love to have a test on a synoptophor but alas, in the UK opticians are private practices primarily concerned with making spectacles and screening for problems. As soon as they find a convergence problem it would be off to see an optometrist (don't know if this is the same wherever you come from?) under the control of the National Health Service. Since I am only interested in my eyesite and do not have a real problem I couldn't expect to, and wouldn't be, refered to an optometrist to be tested. I could probably pay for a private consultation, but I'm sure that would be serious money - and I'm not that interested!

You can just come to Poland, here you would pay not more than 30 GBP for that kind of service. I think the trick would work with every country which is futher away in east direction than Germany...

I was hoping to do things properly with some polaroid filters on my camera and under my enlarger

That's not it, film does not "see" polariation of light. You need a stereo camera, which records two normal images from slightly different positions/angles, and then the images would be *projected* through polaroid filters and watched with polaroid filters. So basically you need a stereo camera, projector and two sets of filters.

 

[removed account4]

there have been stereo cameras around since photography was born.
i don't think it would be hard to find a stereo-opticon and some "vintage" stereo cards
through a store, or ebay ..

http://www.stereoscopy.com/
http://www.berezin.com/

sell the equipment to make your own too

it isn't hard, even with a normal camera. you just need create a bracket to allow your camera to slide over 4-6" .
take 2 images of the same view. there are oodles of 120 format stereo cameras, some expensive, some
are like box cameras. the lomo/sputnik ones from what i have read suffer from light leaks
holga makes a stereo camera that about 70$ usd as well.

this person has a great collection!

http://www.ignomini.com/photographica/3dcameras1.html

good luck!
john

 

[John Koehrer]

There are several on f295 making stereo images with pinhole cameras.

 

[David William White]

I've got a Stereo Realist. A fellow APUG member twisted my arm to get one. They seem to be built to last, and mine works like a peach. There are places on the internet that supply the slide mounts and viewers, so it's very much alive. I shoot colour transparency, but you could load up with B&W and reverse process it. That is, bleach after development but before fix, then re-expose, develop, and fix. Then cut.

John mentioned a slider bracket. Just one limitation on that is the scene can't be changing between exposures. I like the stereo camera because I take street photographs with lots of things going on. They are a total treat.

 

[Mike1234]

Anyone want to buy a pair of ARSAT 37mm fisheyes from the same factory run? You could build a stereo 180 fisheye full circle camera using two sheets of 4x5 or a single sheet of 4x10 or just shoot 8x10 and crop.

 

[steven_e007]

You can just come to Poland, here you would pay not more than 30 GBP for that kind of service. I think the trick would work with every country which is futher away in east direction than Germany...

If I ever do... maybe I can look you up?

That's not it, film does not "see" polariation of light. You need a stereo camera, which records two normal images from slightly different positions/angles, and then the images would be *projected* through polaroid filters and watched with polaroid filters. So basically you need a stereo camera, projector and two sets of filters.

Yes, you're right. I was expecting either two cameras, or a sliding camera backet, or making a stereo camera - but yes, the filters would be on the enlarger lens and in the glasses. Still not sure of the exact proceedure, but maybe best to start experimenting....

 

[steven_e007]

Can I ask... are you colour blind? That would impact your results with the colour 3D tests.

No... I'm not colour blind. My eyes are pretty normal. My left isn't quite as good as my right, but good enough to read with. I don't wear glasses and on the whole my eyesite is as good as it gets for my age - it's just I don't have the capacity to see 3D. Or at least, most forms of 3D, as I said, black and white through polaroid lenses sometimes works, for some reason. I haven't tried coloured glasses with BW images, but I suspect they won't work. I'll have to try. My hunch is that the difference may be that rods (used for BW vision) are scattered more widely around the retina. Cones (colour) are concentrated in the fovia (the very central bit). I wonder if when I combine images from my two eyes my brain 'cheats' by taking the central fovia from just one eye, ignoring the other fovia, but combining the (easier) periphery from both eyes? Therefore BW is easier because the fovia is not so essential to BW viewing?
Don't know - just guessing!

 

[HMFriedman]

Can I ask... are you colour blind? That would impact your results with the colour 3D tests.

Color blindness does not impact the results of color generated 3-D images. The color filters, in combination with appropriately colored images, serve only to isolate the left and right eyes' images, so they can be manipulated and then re-combined into a single (hopefully) 3-D percept. This works in the same way as your colored camera filters do, and is independent of the individual's ability to discriminate different colors.

 

[edp]

there are oodles of 120 format stereo cameras, some expensive, some
are like box cameras. the lomo/sputnik ones from what i have read suffer from light leaks

I have a Sputnik, and while it's certainly no luxury camera, I still use it more than any of the others in my bag. It's brilliant.

A "new" Sputnik needs some attention to stop it leaking light, but it's not much more complicated than glueing a length of wool into the groove that forms the light trap. It's shiny inside, so it also needs some kind of non-reflective material applied to the inner surfaces. After that, and the CLA that any fifty-year-old camera needs, it's a decent 120-format stereo camera for a fraction of the cost of the modern Chinese equivalent.

 

[removed account4]

I have a Sputnik, and while it's certainly no luxury camera, I still use it more than any of the others in my bag. It's brilliant.

A "new" Sputnik needs some attention to stop it leaking light, but it's not much more complicated than glueing a length of wool into the groove that forms the light trap. It's shiny inside, so it also needs some kind of non-reflective material applied to the inner surfaces. After that, and the CLA that any fifty-year-old camera needs, it's a decent 120-format stereo camera for a fraction of the cost of the modern Chinese equivalent.

i have been sooooo close to buying one of these, and /or a gaumont ...

i hate threads like this, cause i always want to buy more stuff ...

its a good thing i'm flat broke eating lentils in a refrigerator box leaching
wi-fi from the guy next to me in a double-wide ( washer dryer box ) ...

john

 

[steven_e007]

Ok...

I haven't done any experimenting yet,

but I have been thinking.

Taken some paracetamol, now, having a lie down....


Thing is - I stillcan't see how to make a polaroid stereo image. As was pointed out further up the thread, BW film isn't polarization sensitive - but then neither is the paper. So... projecting two seperate images on top of each other fromthe enlarger through polaroid lenses can't work either.

So... how to encode a single image so that when viewed through glasses with differently polaroid lenses, you see a 3D image?

That is the question!

Any ideas?

 

[Q.G.]

You can't. Nothing is encoded in the recorded image itself, except the difference in perspective between the left and right eye images.

The polarisation-separation works, because the light used to view the images in is polarized. 90 degrees different for each in the pair. View those through polarizers, also at 90 degree angles, and each eye will only see one of the two.

You can't turn a single image into a stereo pair.

 

[Marizu]

Our stereoscopic vision works because both of our eyes see slightly different views of the world.
Stand a friend in front of you and close one eye. Then open it but close the other.
Without getting into how it works, the polarised glasses make it so that one of your eyes sees one picture and the other eye sees another picture. Born of these pictures are usually taken with the camera about the width of your eyes apart.
I struggle with anaglyphs and the coloured glasses but when one eye sees one picture and the other sees another stereo works great for me.
There are a few ways that you can make your own but you will need some kind of viewer.
One way that has been suggested is to take two pictures of the same non-moving scene with one camera. Make the pictures 6-8 cm apart. You will probably need to use a tripod.
If you shoot on slide film then there are a number of viewers and mounts that you can buy.
I use a viewer that came with a book of 3d photos from the 1840's called "A Villiage Lost and Found." You can mail order this from Amazon for about £20. There is also a set of cheaper books that I saw on there that have a slightly different viewer that is attached to the book. I think the one that my mate has is called "London in 3D" or sonething. It is also full of pictures from the 1800's.
Both books have lenses in the viewers that somehow make it so that one eye sees one image and the other eye sees the other.
You just need to print out your images so that they are similar size (around 6x7cm) as the ones in the book, stick them under the viewer and then it should just work!
They can be really impressive.
At the moment, I am using an arrangement of mirrors to view 20x16" stereo pairs of silver prints from 4x5 negs. They are working really well.
If you are prepared to resort to digital, you can test this by taking left side photo followed by right side photo and then print them in Boots (UK high street chemist) on 6x4 mini prints with either two images or four images per page. Pop them into one of the viewers that I mentioned and they work in 3d.
Good luck. There is a lot of fun to be had!

 

[steven_e007]

Bit of a recap - going back to the original post.

The one type of image I personally can see in stereo is a single BW image, consisting of two super-imposed views from slightly different viewpoints, viewed through polaroid glasses. I have never managed to get two separate images (using a Victorian style viewer) to work.

Usually, combined single images are in two colours and are viewed using two different coloured lenses - so you print two views on colour paper. Fairly easy. But how on earth do you encode two images on one sheet of paper with different polarisation?

I'm beginning to think you might need to make a sandwich of BW slides with polaroid filters... but surely the top filter would block all the light from the image/filter underneath?

 

[Rudeofus]

But how on earth do you encode two images on one sheet of paper with different polarisation?

That's hardly possible.

I'm beginning to think you might need to make a sandwich of BW slides with polaroid filters... but surely the top filter would block all the light from the image/filter underneath?

No, don't make a sandwich of slides. What you need to do is use two slide projectors and somehow put a polarizer in front of their lenses. Make sure the filters are 90° turned with respect to each other around the filter axis. With a setup like this you basically project one slide with some polarization, and the other slide with the orthogonal polarization.

If you now hold a simple polarizer filter in front of your eyes, you can see the projected image of either one or the other slide. Now make glasses where each eye sees only one polarization and you're set.

 

[steven_e007]

That's hardly possible.

It must be possible, because I've seen it!
There is a type of test chart used in opticians which uses polaroid spectacles - and for some reason it is usually an image of a house fly... I can't workout how they do it, though...

No, don't make a sandwich of slides. What you need to do is use two slide projectors and somehow put a polarizer in front of their lenses. Make sure the filters are 90° turned with respect to each other around the filter axis. With a setup like this you basically project one slide with some polarization, and the other slide with the orthogonal polarization.

If you now hold a simple polarizer filter in front of your eyes, you can see the projected image of either one or the other slide. Now make glasses where each eye sees only one polarization and you're set.

Yep... I can see where you are coming from. That is an interesting idea. I bet it would be pretty tricking to align the two projectors exactly, though...?

 

[Rudeofus]

It must be possible, because I've seen it!
There is a type of test chart used in opticians which uses polaroid spectacles - and for some reason it is usually an image of a house fly... I can't workout how they do it, though...

Interesting, never seen this. Is this a printed test chart or some backlit slides?

Yep... I can see where you are coming from. That is an interesting idea. I bet it would be pretty tricking to align the two projectors exactly, though...?

It should be pretty easy since both projectors can be placed close to each other and you can use some edge in the picture to help you with the alignment. As long as you don't wear the polarizer goggles, you just see two images projected on top of each other.

 

[steven_e007]

Interesting, never seen this. Is this a printed test chart or some backlit slides?

.

Hi....

No, nothing is backlit - it is just a flat sheet. I'm beginning to think it is maybe some clever sandwich of layers and filters or something - but I can't see how it is done.

Here is a link to one:

http://www.eyesfirst.eu/epages/61989410.sf/en_GB/?ViewObjectID=5973646

The test I used wasn't a titmus fly test - it was a different make, but still a fly (seems flies are traditional for stereo testing!). The titmus is mounted in a folder, so could be quite a thick sandwich of something, but the test sheet I saw was just like a laminated photograph - really not so thick.

 

[Rudeofus]

Wow, learn something new every day ... apparently these test sheets use a vectograph for creating this effect. Not sure how to make such a thing at home. Note, that you can not replicate this with regular slides and polarizer gel!