Autostereoscopic multidimentional processes and Lippmann's Integral Photography

[Lionel1972]

The Lippmann's plates thread ( (there was a url link here which no longer exists) ) got me interested in Integral Photography and autostereoscopic processes in general.

I recently came across a magnificiant example here:
http://www.flickr.com/photos/garyauerbach/5068112016/

The photographer used a personal process that seems connected to Lippmann's Integral photography. He told me that he can be contacted about this through e-mail.

Gary Auerbach's email : platinumphotographer@yahoo.com


More on Integral Photography here:
http://en.wikipedia.org/wiki/Integral_photography

Anyone else ever tried autostereoscopic processes ?

 

[platinumphotographer]

Hi Lionel1972...

Thanks for the post about autostereoscopic multi-dimensional platinotype's using lenticular screens.
Best wishes to all at APUG. Better email is garyauerbach@hotmail.com

Gary

 

[holmburgers]

Pertinent Links & Resources

Overview and history: http://www.outeraspect.com/history_lenticular.php

This company can supply lenticular sheets and fly's eyes array: www.fresneltech.com

This PDF... http://fresneltech.com/pdf/FresnelLenses.pdf is a catalog. I contacted the comany a while back and they said, "#300 has been used to make 3D pictures, and the pictures were good, but somewhat coarse. You might also consider the #360. You can ask sales@fresneltech.com for a quotation; our catalog prices are not current, I am sorry to say. Sales should also be able to supply data on the lenslets. Please be sure to request a thickness such that incoming light is focused on the plano side of the arrays." Minimum $100 order.

These would be suitable for integral photographs; I've not checked into lenticular sheets specifically.

Takanori Okoshi's "Three-Dimensional Imaging Techniques" is an excellent book on the topic. Most university libraries should have a copy. Half is hardcore geometry but the other half is descriptive explanations that are suited to the lay-man.

Lippmann's Original disclosure in English: http://people.csail.mit.edu/fredo/PUBLI/Lippmann.pdf

http://www.tgeorgiev.net/Lippmann/ (more, as well as French version)

Dead Link Removed (English & Russian texts)

This is an abstract only: http://www.opticsinfobase.org/abstract.cfm?id=16069 But an alternate method that doesn't use a screen per se, but rather thousands of tiny glass beads. This is explained briefly in Okoshi's book. It is a route to consider, but poses difficulties. However, retro-reflective beads used in road paint and street signs are theoretically capable of meeting the necessary conditions. These can be had for next to nothing from hobby suppliers on eBay and the like. Worth noting.

At some point in the 80's (IIRC) Roger de Montebello owned/operated a company in Ann Arbor, MI called MDH Products. They created "Integrams" which were non-pseudoscopic integral photographs. Very little information can be found about this; but supposedly they were quite good, perhaps the best examples. Please contact me if you have any information about this. Patent: http://www.google.com/patents?hl=en...ebello"&printsec=abstract#v=onepage&q&f=false

Lastly, I was going to post a link to the Holography Forums, which had some discussions and a few posts with examples. However, follow the link and read the bad news for yourself (site shut down due to legal troubles). http://www.holographyforum.org/

 

[Hologram]

Lastly, I was going to post a link to the Holography Forums, which had some discussions and a few posts with examples. However, follow the link and read the bad news for yourself (site shut down due to legal troubles). http://www.holographyforum.org/

The holographyforum is about to be re-constructed, see: http://holography.no-ip.org/

 

[holmburgers]

Gary's Method

Hey there,

So I actually talked to Gary Auerbach on the phone last night and we had a nice discussion about his method for making this 3D picture, among other topics. He used a camera with a motorized front lens board that shifts from side to side during the exposure. This, coupled with a lenticular sheet that is pressed in contact with your film will result in an "encoded" image on the film. Then, once you reversal-process it or contact print it, you place an identical screen back in registration with it to "decode" it and boom you've got a 3D image.

I believe that all the examples of lenticular prints that you find from the 70's (you know, lap-dogs, bouquets of flowers; the stuff you find at antique malls) were made with this kind of camera. It gives a continuous stereoscopic image within a limited range. I have an example and it is quite good, but as you go too far to the left or right, the image kind of resets, that is, starts over and there's a weird point where this discrepancy is very distracting. I believe this is called "flipping", in the words of Takanori Okoshi. However, this is different from 'flip-animation' lenticulars, which are meant to show an animation of sorts.

Amazingly, here is the camera design that he used, a WT-102 Multi-dimensional camera. I have contacted this lonely APUG'er, hopefully he is still around...

(there was a url link here which no longer exists)

Here is a short discussion of analog lenticular printing. In this description I described a different method that in theory should work, but now knowing about this camera.. it makes a lot more sense today.

(there was a url link here which no longer exists)

Lionel, thanks for starting the thread; I hope there is interest from others as well!

P.S. I'm glad that the holography forum is back

 

[holmburgers]

Check out eBay item #260636497769

Surprisingly insightful description... amazing camera

 

[goahead]

holoforum

Hello,

The succesor of the holographyforum.org has it's own domain now and is accessible at http://www.holoforum.org/forum

The postings from the old forum should get available soon.

 

[holmburgers]

The other day I saw an Avatar pinball machine (silly, I know) that had an auto-stereoscopic backglass. It was obviously a digital print, but it had excellent resolution and I'm pretty sure that it had up/down and left/right dimensions. That means it couldn't be a simple lenticular print, and must have been something like a fly's eye array. I tried to look at the screen, but it was very fine and I couldn't see any small detail that would tell me what kind of screen it was.

But I've seen a lot of these lately, at movie theaters, in various advertising things, etc. Digital printing companies out there are doing very impressive work with auto-stereoscopic prints. Doing things digitally certainly makes it a lot easier, but it's not to say that these methods couldn't be done analog.

Still interested in this stuff, how about you Lionel?

 

[holmburgers]

Beaded Plate Recording of Integral Photographs
http://www.opticsinfobase.org/abstract.cfm?uri=ao-8-11-2329

I haven't read this, but I'd like to. You can make a fly's eye array from small glass beads coated on a photo plate. I've read that the balls should have a refraction index of n=1.0*. Retroreflective glass beads, which are cheap as dirt for crafts, or slightly more expensive for street signs, road paint, have indices approaching this. Therefore, they should theoretically work.

The trick would be to coat the film/plate with these beads. I think that the authors of this paper wetted a glass plate in photo-flo and poured the beads on it. The gelatin got sticky enough to accept the beads and despite being exceptionally fragile, it should work. However, since glass plates aren't really practical these days, unless you are coating your own emulsions (lightbulb anyone!??!) it'd be nice to make it work with film.

I'm wondering if a thin solution of gelatin would be permeable enough by processing solutions to be successful. You could mix the beads and gelatin and coat the film.

Then, you'd expose the plate without a camera; each bead acting as its own lens. Not sure what kind of "focusing range" one would have, but it'd be fun to figure out.

Take this idea one step further, you could use red, green and blue colored glass beads. I don't know if they exist with the proper retroreflective properties, but such beads do exist for craft purposes. Imagine it though... you could have an integral color photograph a la screen-plate photographs. From a black and white film and no camera, you could get a 3-dimensional, color image.

*edit: Ok, thanks to my girlfriend being a student, I got the article. PM me if interested. Turns out the refractive index is 2, not 1. "The glass balls have a refractive index close to two and act as little lenslets. Their focal length is equal to their diameter and they therefore form an image on their back surface."

 

[Photo Engineer]

Kodak has a patent on a multilayered 3D moving lenticular system. I have seen 2 examples of it used in movie posters. One, for Jurassic Park showed a moving T-Rex that seemed to leap out at you as you walked by, and the other was for STTNG-Generations in which the Enterprise seemed to fly through a Star Trek logo as you walked by. They were made available to the movie industry but never caught on.

Kodak published their annual report one year with a 3D lenticular photo on the cover.

Kodak also made a 3D lenticular print material to make 3D prints from the Kodak Stereo Camera, but none of this ever caught on.

It was all quite good, but perhaps before its time.

PE

 

[holmburgers]

Indeed. I think it is coming into its own, and I personally think that the 3D TV's of the future will/should use integral techniques. Wearing glasses is not sustainable, and if you can program the pixels behind the fly's eye elements, then you've got glasses-free 3D. With HD resolution, this shouldn't be too difficult. And such a system was demonstrated in 1978 and some modern prototypes from Toshiba, etc., utilize this idea.

PE, this article mentions the use of Kodak 649 GH plates and Kodak 649F plates. What are they exactly? Not still available I assume.

 

[Photo Engineer]

Sorry, I have no idea. I have never worked with plates at EK.

PE

 

[AgX]

Those were Lippmann type films and plates, used for spectroscopy and holography.

 

[holmburgers]

Open Solicitation for Information Regarding Integrams and Roger deMontebello

http://www.invention.smithsonian.org/resources/MIND_Repository_Details.aspx?rep_id=1834

This inventor, Roger deMontebello, is someone that I'd like to know more about. He resided in Ann Arbor, Michigan in his later life, as I understand and you can see his list of contributions to imaging in this link.

Specifically he made Integrams which were, as I've heard/read, perhaps the height of integral imaging. That could be an overstatement, but they were not pseudoscopic and apparently had a large viewing angle. There was some kind of commercial aspect to the invetion, so there has to be examples floating around out there.

I believe the integram was sold/developed under MDH Products Corporation of Ann Arbor, MI. In short, if anyone has information regarding Mssr. deMontebello or has an integram or anything related to integrams, MDH products, etc. please contact me!

The information out there is scarce to say the least, and it deserves to be studied and made available to others. I would pay good money to get my hands on an integram.

 

[holmburgers]

Auguste Lumière's Photostereosynthesis

This is an interesting and crude way to get a 3D image. Not quite up to the standrards of integral imaging, but interesting in it's own way and perhaps easier to do.

These links will give an idea:
http://www.conceptlab.com/photostereosynthesis/cameracontroller.html
http://www.conceptlab.com/photoster...synthese-academie-des-sciences-nov8th1920.pdf
http://www.shardcore.org/shardpress/index.php/category/photostereosynthesis/

The idea is to take several pictures with incredibly shallow depth of field of a subject at close range, each shot focusing on a different plane. You then have transparent prints made, slides, and you stack them with separation between each other. The 3D illusion is achieved by the in-focus areas and our eye's propensity to disregard the fuzzy, out of focus areas.

The links are from two different people; the first guy who has a lot of good historical information and a scheme for doing this with a modern dSLR.

The other guy has some examples that he actually made. Neither are groundbreakingly good, but I think there is a lot of potential here. By eliminating the out of focus areas in each shot, perhaps a cleaner 3D view could be achieved.

One thing that I'm not certain about is the necessity of changing the focal length. I need to keep reading perhaps, but I'm not sure how this is achieved exactly, especially with the 50mm 1.2 that the author proposes to use. Perhaps reading the original paper by Auguste would shed some light (2nd link) on the matter.

 

[holmburgers]

Rough OCR and Translation - But legible

Here is the 1920 Lumiere paper on Photostereosynthesis, OCR'd and roughly translated. Came out pretty good, but could certainly use some cleanup. Just wanting to make this stuff available to people...

BTW; I used these to programs to do it:

Take a screen shot of the PDF, crop in Paint and upload to this website -> http://www.onlineocr.net/

And translate with babelfish -> http://babelfish.yahoo.com/

 

[Mustafa Umut Sarac]

Chris ,

Do you aware of rotating mirror 3D 360 degree viewer ? You can walk around it and its like real. I think computer modellers would be crazy about it. I think finding a 360 degree 3D camera technology or a cheap viewer technology could make money. Viewer and camera could work with 2D slices and this is not too difficult may be.

Best ,

Umut

 

[holmburgers]

I don't think I've heard of that specifically, but just today I've found many similar schemes that utilize rotating helixes. Roger deMontebello invented one himself, that's how I found it. (http://www.freepatentsonline.com/3428393.pdf)

Do you have a link perhaps, or a reference?

And look, do you remember the 3D television, transparent cube idea? Well, as usual I wasn't the first! -> http://adsabs.harvard.edu/cgi-bin/nph-bib_query?1994SPIE.2177..196M&db_key=INST

EDIT: aahhh, here check out these schemes -> http://www.felix3d.com/web/index.php?show=Hall of Fame&type=2

UPDATE: Crazy patent: photo sculpture http://www.freepatentsonline.com/2891339.pdf

 

[Photo Engineer]

There is a recent patent by Paul Gilman for the rendering of 2D motion picture in full 3D. It works very well.

PE

 

[Mustafa Umut Sarac]

Ron ,

Is that Paul B Gilman ? There is several patents at google patents related to Kodak but I couldnt located one in this technology even after research at google.

Can you describe little bit ?

Hey , do all knows Kodak invented first digital camera and I think its their natural right to gain in this.

And 25000 dollars Leica S2 Body uses Kodak Sensor costs 12000 dollars to them

Umut

 

[Photo Engineer]

Umut;

Paul got this patent about 3 years ago and it would be in the last 3 or 4 he got. Basically it works as follows using Frame number and Eye as 1L 1R etc. Each full picture is separated by a semicolon. The frames are projected through polarized lenses.

1L, 2R; 2L, 3R; 3L, 4R; 4L, 5R; 5L, 6R etc. The slight motions of the camera and actors will supply the viewer with 3D images.

Another method is based on size in which the left and right eye see different size images and a third introduces prisms to cause pseudo color based 3D. These latter are well known and have names. I posted the names here in a previous thread similar to this one. The size based pseudo stereo vision is often seen by those having had cataract surgery in one eye. The new lens and glasses render that eye image at slightly different size. Those with problems such as Exophoria (SP?) and who therefore use prisms often see stereo images based on color with colors stacked in "layers" with for example a red appearing closer than a blue.

PE

 

[Mustafa Umut Sarac]

Thank you Ron ,

This is similar to softwares extracts 3D from motion film.
I had been posted a idea several years ago to here as a technology which extracts 3D from motion film.
It was based on similar idea and comes from softwares.
May be they stealed my idea , this was not first and will not be the last.
Some of my patent applications returned from us with comical reasons , they say they could not scan my pdf files because there were not enough spaces between words ! They requested 400 dollars from me to correct it.
I am sure someone from defense industry would be eat my pie.
No body sees , no body hears and patent applications being stealed.

Umut

 

[Lionel1972]

The other day I saw an Avatar pinball machine (silly, I know) that had an auto-stereoscopic backglass. It was obviously a digital print, but it had excellent resolution and I'm pretty sure that it had up/down and left/right dimensions. That means it couldn't be a simple lenticular print, and must have been something like a fly's eye array. I tried to look at the screen, but it was very fine and I couldn't see any small detail that would tell me what kind of screen it was.

But I've seen a lot of these lately, at movie theaters, in various advertising things, etc. Digital printing companies out there are doing very impressive work with auto-stereoscopic prints. Doing things digitally certainly makes it a lot easier, but it's not to say that these methods couldn't be done analog.

Still interested in this stuff, how about you Lionel?

Hi Chris,

Yes I'm still interested in this stuff. I've just been busy lately. I've noticed the 3D posters and flyers here too, along with the new DVD and Blu-Ray 3D covers that seem using lenticular screens too.
Thanks for the Louis Lumière note, very interesting. I'll try to come up with a decent translation when I get some free time. He explains how to get extreem shallow depth of field with a regular fast lens.

 

[holmburgers]

Hi Chris,

Yes I'm still interested in this stuff. I've just been busy lately. I've noticed the 3D posters and flyers here too, along with the new DVD and Blu-Ray 3D covers that seem using lenticular screens too.
Thanks for the Louis Lumière note, very interesting. I'll try to come up with a decent translation when I get some free time. He explains how to get extreem shallow depth of field with a regular fast lens.

Sounds great Lionel! I'm thinking to procure the lenticular screen from some of these aforementioned products; for instance I have a tin of tea that has a lenticular screen... weird! Hopefully it will allow for some kind of testing though.

 

[holmburgers]

http://www.nytimes.com/2011/06/22/technology/22camera.html?pagewanted=1&_r=1&src=ISMR_HP_LO_MST_FB

New camera coming out that uses similar concepts to integral imaging. You heard it first here ladies & germs... "light field" imaging will replace the lens as we know it.