Instructables
An anaglyph is a color image that creates the illusion of "3D" depth when viewed through color filters that separate the left and right views. The image shown here is an example viewable through green/magenta glasses.

Although various color combinations and processing variations have emerged over the years, the basic concept of an anaglyph is largely unchanged  since the 1850s. Nearly all methods start by capture of a stereo pair of images which are then manipulated to create the anaglyph. In contrast, the method discussed here involves modifying a digital still or video camera to directly capture a high-quality anaglyph in a single shot -- with no post-processing needed.

Did I mention that the reversible modification to your camera can cost less than $1?

Step 1: What You Need

The parts you'll need are:

1. The camera and lens to be modified

2. Two pairs of identical paper anaglyph viewing glasses

3. A method of mounting (paper, tape, and scissors; optionally, a lenscap and drill)

This method works with most cameras and lenses, but works much better with some than with others and is a little touchy about some details. Don't be scared by the large number of steps in this instructable -- that's just trying to make sure that you get things working as well as possible without a trial-and-error process. This is easy.

Cost? Well, you probably have most of the stuff you need. The paper anaglyph viewing glasses are widely available for free in small quantities. I've bought hundreds at an average cost of $0.32 each including shipping, which would bring the "new purchase" cost to $0.64 for the two needed... easily under $1. Using a lenscap instead of paper printout for mounting adds about $1 to the cost, but yields a "more professional looking" and more durable device.

Note that the post processing described in steps 12, 13, and 14 is optional. You don't need a computer to make anaglyphs by the method described here.
 
Remove these adsRemove these ads by Signing Up
vishuvishu2 years ago
dear sir
how to use this technic in mobile camera???
please reply me.....
ProfHankD (author)  vishuvishu2 years ago
One posting would have been enough. ;)

Anyway, if by "mobile camera" you mean a cell phone camera, then the quick answer is that the lenses are so tiny on most that it is not easy to do this and the effective stereo baseline is so small that only macro images would show good depth. That said, try making the apertures using a pin -- that has worked on a few lenses for which I needed very small, tightly placed, apertures. You can still use the anaperture tool (see step 5) to design the stop and use bits clipped from gel filters to impose the colors.
vishuvishu2 years ago
dear sir, how to use this technic in mobile camera???
please reply me.....
vishuvishu2 years ago
dear sir
how to use this technic in mobile camera???
please reply me.....
vishuvishu2 years ago
dear sir
how to use this technic in mobile camera???
please reply me.....
iectyx3c3 years ago
5 stars. The best.

Would you post a wiggle-gram of one your images?

For other readers:
http://www.instructables.com/files/deriv/FI7/S7I6/FUPUPGG3/FI7S7I6FUPUPGG3.MEDIUM.gif
ProfHankD (author)  iectyx3c3 years ago
Well, high-quality conversion of an anaglyph into a wiggle-gram is a research topic, not a standard technology. However, I understand that without the right glasses you can't see how well this works. Fair enough.

Here's a B&W wiggle-gram constructed using GIMP to separate the left and right sides, convert each to B&W, and then combine them as a 250ms-per-frame looping animated GIF. The result is slightly wrong due to the different color sensitivities of the left and right images within the anaglyph, but it is directly viewable....
an14.gif
This looks great, but I think it would look better if the delay was short. Maybe 33ms?

Can't wait to try this.
Wow! Thanks for the wiggle-gram.

This really shows how brilliant your single lens anaglyph method is.

Absolutely incredible to realize that the different sides of a lens produce different angles of view on the film or sensor.

Just a superb i'ble.
That gave me a headache, but was so worth it. really cool!
Hi My main concern with this is the quality of the image.

But after uploading an image myself I see that what you see here doesn't clearly indicate the quality of the original, so maybe it works better then I thought.

This image was only an 800 kb file, but looked very clear before I uploaded it without the artifacts around the edges that I see in it now.

I wish i could see a real example of what the flag image looks like.

This is my Character Lucy, the star of my (in my opinion) humorous videos, shown in 3D.  My goal is to do a Lucy video in 3D.  I have to come up with a 3D joke.

Here's a link to the original.
http://i1238.photobucket.com/albums/ff491/nohjekim/Greetings2B.jpg
Greetings 2B.JPG
Hi

Another way to do this that give really high quality results is to make a device that fits on you tripod that allows your camera to slide sideways 3".

I just made a block of wood with a slot in it that fit another block that I mounted my camera on, using the normal tripod mount screw hole.

I had a stop on each end that allowed the camera to slide 3".

Once you have both images shot process them in Stereo Photo Maker (a free downloadable program) and you will get a very good Anaglyph image that you can view on your computer monitor.

The drawback on this is that you have to shoot things that are not moving, since the two images are not shot at the same time.

But if you have a workshop you can probably do this with stuff you have laying around.

ProfHankD (author)  MikeHawthorne3 years ago
There are many ways to capture depth information, and directly capturing a stereo pair with a slide mount as you suggest is very popular. One advantage is that you can make the separation between views larger to enhance the stereo effect. The catch is that the camera would need to make two exposures to post-process into an anaglyph and the slide mount (and tripod) tends to be bulky.

Directly capturing anaglyphs as described here only requires carrying a little filter that you can mount/unmount as desired and you can see the anaglyph directly on the camera's display, even in live view.
Dear professor,
Most of my glasses are of red/blue variety, they are good for most of the anaglyphs pictures and movies on the web. Would a red/blue combination work? Or it is worth it to stick to your suggested green/magenta combination?
ProfHankD (author)  cafepollution3 years ago
Red/Blue is only usable for monochromatic scenes. However, about half the time I've seen people say Red/Blue they actually mean Red/Cyan....
ProfHankD (author)  cafepollution3 years ago
Red/Cyan works too. The theoretical problem is that the Bayer pattern imposed on the sensor in most cameras gives a 1:3 pixel count imbalance between left:right views for Red/Cyan. Fortunately, this resolution difference is not obvious under typical viewing conditions.

To answer your other question, I used a Sony A350 DSLR to capture the anaglyphs in steps 12 and 15. In general, DSLRs -- and film SLRs -- with appropriate lenses produce high-quality anaglyphs with greater depth than most smaller format cameras. However, optical viewfinders can be too dark for good composition, so a live view that "gains up" nicely can be very useful.
Sorry I forgot to ask - would this work with SLR cameras e.g. hacking the lens cap for Nikon D40? Thanks.
rimar20003 years ago
Sorry, I hate to be disagreeable, but I think to get a 3D image is essential to have a distance between the objectives of at least 6 cm. I make anaglyphs since years ago, using various methods, and are much more 3D than these photos. You can see some of them here at Flicker. Not all of them are well done, they are a bit old.
ProfHankD (author)  rimar20003 years ago
Shorter baselines do better with close-up subjects. I've used MANY different technologies to capture depthmaps or even true 3D scene models, and every method has good and bad points. The good here is the ease, low cost, lack of sync/exposure issues, and quality of optical matching between views.
Yes, you are right about close-up subjects. Also that each 3D technique has some bad points.

That I say is this simple and cheap method is only that: simple and cheap. But the results are not good.
This looks cool. Totally want to test it out soon. We have a laser cutter here, after all