JOHN'S 3D GUIDE

(Last Update November, 2004)

CHAPTER 1:    BASIC PRINCIPLES OF 3D PHOTOGRAPHY & THE SLIDE-BAR

Stereo photography was very popular in the 1950’s.  The medium fell out of favor for a long time, probably because 3D images were considered somewhat difficult to view, and projection (perhaps the best way for large audiences to “see 3D") requires careful registration of the image pairs.  However, with modern equipment an impressive presentation is relatively easy to put together.  More importantly, true 3D viewing (as opposed to 3D rendering in computer graphics) is being supported by an ever increasing number of big-time computer equipment and software vendors. There is now a wide range of special graphics cards and visualization glasses that make digital "virtual reality" a reality indeed.  As a result of this, and  because the medium is just so jaw-dropping impressive, stereo photography is making a comeback!   Stereo photography clubs and associations are seeing a resurgence of interest, and websites containing 3D images abound.   For example, a wide selection of 3D images are shown at our own stereo gallery  , where viewing tips are discussed.

These basic instructions were originally put together to encourage  active outdoor sports enthusiasts, and others as well, to try 3D.  The methods described here apply equally to stereo presentation of scientific imagery.  This write-up is introductory.  More exhaustive references are cited at the end of this chapter.

What equipment does one need to get started? 

Actually you can start with just a single camera, film or digital, high or low resolution.  This chapter describes the basic configuration, a simple slide-bar setup for use with a single camera, and concludes with a brief overview of dedicated 3D cameras.

THE BASIC IDEA

High quality stereoscopic pictures can be made with one or two cameras.  In either case a "stereo pair" consisting of a single left (L) and a single right (R) picture, properly mounted and aligned, are obtained, simulating the action of the left and right human eyes.   To facilitate the acquisition of the two images, the camera(s) should be placed or attached to an alignment bar that will minimize pointing errors. 

The images are taken with the cameras are spaced apart, or the single camera moved laterally,  by a distance known as the "stereo base". 

Fig. 1.1.  STEREO BASE For illustration we show here two cameras mounted on a bar.  Usually one camera is fixed and the other can slide sideways and then be clamped.  Alternatively a single camera may be slid sideways between pictures.  It is important that the cameras face straight forward and are not toed in.

Note that in figure 1.1 the cameras' lines of sight are parallel, not angled in to a central focal  point.  For still photography this avoids opposing keystone effects for the L and R cameras that you will get if the cameras are toed in.  However,  the the horizontal shift between cameras means that some portion of the right edge of the right frame, and of the left edge of the left frame,  will not have overlapping information from the other camera.  These areas have to be cropped off.  This is part of the important stereo image alignment process, registering the stereo window, that is discussed in chapter 5.  

Instead of using two cameras, a single camera can be employed for certain types of photography in which the subject does not move significantly.  In the above illustration, the single camera technique means that you take a shot with the camera in the right position (R).  With the bar mounted on a tripod to keep it from moving, the single camera is then moved sideways (or remounted) to the left position, whence a second (L) image of the pair is made.  Obviously the subject must not move between shots.  Water flowing or clouds drifting fast won't work.  But this technique is effective for rocks, plants (with no wind), and other "still-lifes."   Some photographers have success with this "Cha-Cha" method by simply hand-holding a single camera and shifting position slightly (usually a few inches) between shots.  A simple Single Camera Slide-bar is described below, and the two camera Twin-Cam is discussed in depth in Chapter 3.

THE STEREO EFFECT

How does two-shot stereo work?   The stereo pair  is usually made up of two slides or digital images (L & R), or of two prints mounted side by side.  

Fig 1.2  MAKING A STEREO PAIR Here is an example of a scene with a big fir tree in the foreground and a red steeple in the background.  The two cameras (or the single camera in the slide-bar case) produce an image pair in which the fig tree appears off to the right in the L image, and in which the fig and the steeple are lined up in the right (R) image.  The latter situation comes about  because the fig and steeple lie on a straight line angling out from the lens of the right camera (see thin red ray-trace lines in the illustration).   IMPORTANT:  Imagine what happens if the stereo base is increased by moving the left camera further to the left.  The fig tree moves slightly to the right in the L image, and the steeple moves more to the right.  If the steeple were infinitely far away, and the left camera is slid to the left without any rotation, then the image of the steeple would not move on the image much at all.

Now let us notice what happens when we view the image pair.  Let us assume that we can arrange to view the pair more or less as they are shown in Fig. 1.2.  That is, imagine that our left eye sees image L and and our right eye sees image R, straight on.  This might be accomplished in a print viewer by placing a small print of L just in front of your left eye, and a small print of R straight in front of your right eye.  Alternatively, imagine placing these in a slide viewer (Chapter 2) to accomplish the same thing.  

Fig. 1.3  REALIZATION (Note:  Here the vertical displacement of the red and green objects in the cartoon is increased for clarity of presentation.) Right eye looks at R objects. Left eye looks at L objects. The "Apparent Image" (in your head) forms where the rays from each eye, that go through the same physical points in the images cross.  So the: Fig trees top appears to be near point N. Steeple appears to be far away (its peak is at point F).

That's all there is to it.  As a little exercise, imagine what happens in the case where the stereo base is increased by shifting the left camera further to the left.  

Answer:  Remember, increasing the stereo base this way causes the L fig tree to move further to the right, while the L steeple stays more or less at the same point.  As the L fig tree moves to the right, the near point (N) moves towards the viewer, because the ray from the left eye to the top of the L fig tree rotates clockwise as the L fig tree image moves to the right.  In this manner you can see that increasing the stereo base will cause the viewer to experience a stretching of distance for a fixed scene.  

HYPO AND HYPERSTEREO

From this we learn that the stereo base is one element in defining the perceived depth.  Another factor is image magnification.  In hypostereo the stereo base is less than the normal eye separation of 65mm (or about 2.4 inches).  In hyperstereo the stereo base is more than normal eye separation.  In ortho-stereoscopic viewing the base separation and the focal lengths of the lenses (43mm = eye equivalent) are set up to reproduce what the viewer would see in reality.  In an "orthostereo display", the taking and viewing lenses are usually the same.  The two 35mm cameras mounted as shown at the bottom of figure 1.1 have a typical separation of about 3 inches.  This is slightly hyperstereoscopic and may be just noticeable in normal viewing.  With a large hyperstereo base distance a person will appear taller or more leggy than in actual fact.  Sometimes hyperstereo is a good thing because it exaggerates the stereo effect!  On the other hand it can render near objects oddly.  For most scenes and action shots where the subject is not too close, the slightly hyper configuration at the bottom of figure 1.1 is fine.

Hyperstereo (extended stereo base) is particularly useful in shots where the nearest object is far off.  Human's do not notice depth on objects more than a couple hundred feet away because the rays coming to the eyes from such a distant object are essentially parallel.  However, by moving the "eyes" (i.e. the cameras) further apart, such rays now come in at differing angles and depth perception is restored.  For example, if the nearest object is 100 feet away, a hyperstereo base with the cameras separated by 1 to 2 feet will render significant depth to the image.  Cameras three feet apart will yield stereo pairs that show some depth even at subject distances of 1000 feet or more.

IMPORTANT:  When viewing objects as illustrated in figure 1.3, if the fig tree separation between L and R increases without limit, the left and right eyes will be looking at such disparate scenes that the brain won't be able to "fuse" the image into a rational scene.  Trying to fuse an image with too much near vs. far separation produces severe strain, nausea, even migraine headaches.   Thus there is a fundamental guideline of 3D photography:

THE BASE SEPARATION FOR A NORMAL LENS (i.e. 50mm on a 35mm camera) SHOULD NOT EXCEED 1/30 THE DISTANCE TO THE NEAREST OBJECT IN THE FIELD OF VIEW.

Examples:

a)  Nearest object 20 feet.  Maximum separation 20/30 feet ~ 8 inches.  

b)  Nearest object 100 feet.  Maximum separation 100/30 ~ 3 feet.  

c)  Nearest object 20 feet but a wide angle 24mm lens is used.  Since the wide angle produces an image where the angular displacements are about half what they are with the 50mm lens (i.e. it is "twice as wide"), the maximum separation is 2*20/30 ~ 1.3 feet.  Alternatively for a 24mm lens the guideline is 1/15.

d)  Nearest object 100 feet, but we want to use a 135mm telephoto to "bring the object closer".  In doing this, image displacements are magnified by 135/50 = 2.7.   Thus the maximum separation is about 20/(2.7*30) feet or approximately 1/4 foot ~ 3 inches.   

e)  You don't always want to push the limit on the 1/30 rule.  Experimentation will show the way.  Telephoto lenses seem to magnify all the potential errors (parallelism in pointing for example), while wide angles generally are more forgiving and better for 3D because they don't compress distance and do not exaggerate pointing errors.

NOTES ON THE 1/30 RULE:

Some experienced 3D'ers consider The 1/30 Rule to be more of a myth than a rule.  In the Art of photography, there are few canonical rules.  1/30 is a good guideline to start with, however.  Exceptions abound in reality.  For example, if the near point and the far point in the scene are close in distance (for example if the far point is not at infinity, but only 10 feet in back of a 10 foot near point), then 1/30 can sometimes be exceeded (maybe to something like 1/15).  In macro work the near points and far points are usually very close (because there is little depth of field when the magnification is high).  Nonetheless I prefer not to exceed 1/30 in such situations.  The ability to fuse stereo images varies widely from person to person.  This ability also varies with the method of presentation (viewer magnification, slide projection beam angle (or projection lens magnification), distance from the screen, etc.)  If you are going to expose a wide range of folks to your 3D art, it may be best not to go wild with the separation.  Two migraines in a large audience are still two to many.  Experimentation is a good way, at least initially.  Bracket (i.e. try the same scene)  with a few different stereo bases and test it out (writing down what you did!).   

HOW TO GET STARTED:

A SIMPLE SLIDE BAR

So you want to try this deal without making a big investment?  Assuming you have either a 35mm SLR or a digital camera you need only to construct a simple slide-bar to test whether you are impressed enough to want to get into this.  While professional slide-bars can be purchased for ~$50 to $100, see http://www.stereoscopy.com/jasper/slide-bars.html), you can make one  for almost no cost at all.

Fig. 1.4  SIMPLE SLIDE-BAR Here are the parts you need. A two foot 2x2 board (find a straight piece) A 1/4 - 20 screw about 2" long. A 1/4 - 20 screw with about 1/4 " of threads (may be cut or ground down). A piece of 1 1/2 " angle aluminum, cut to about 2" - 3" long. A flat piece of plastic or wood about 1/8 to 1/4 inch thick, same length as angle, 1 3/8 wide (to fit on inside of the angle).

Drill a 1/4" hole in the angle aluminum as illustrated.  This will allow you attach the bracket to the bottom of your camera via its tripod socket.

Drill a 1/2" hole in the plastic or wood spacer, with the hole centered over the 1/4" mounting hole.  This spacer is necessary so that the tripod-hole screw head (on the short 1/4-20) will not protrude.  It is "counter-sunk" into the spacer.

Attach the plastic/wood spacer to the inside of the angle using silicone rubber or epoxy.  It helps to lightly sand or steel-wool the joining faces just enough to generate  tiny scratches for the glue to adhere to.  Align the spacer so that the 1/2" hole is centered over the 1/4" hole in the aluminum, as shown in the illustration.

Drill a 1/4" hole at the midpoint of the 2x2.  Use the 1/2" drill to countersink it too.

Fig. 1.5  SLIDE-BAR OPERATION Attach the 2x2 to a sturdy tripod with the long screw.  Some tripods are 1/4-20 threaded.  If not, use a nut. Mount the bracket to the camera base as illustrated. Now you can simply slide the camera sideways between R and L shots, and it will stay nicely aligned, with parallel pointing. You can add a clamp if need be.

To experiment with this rig take it up to a scenic location and find some shots where things don't move.  Slow cloud drift is OK if you are quick between shots (a few seconds, perhaps a little longer if using a wide angle lens).  Trees or flowers moving in the wind won't do.  Maybe a person can sit still long enough (try it).  But this little gizmo works best with a still-life like fruit, houses, street scenes with no moving cars, etc.  For you action guys, cliff-drops are impressive.  The slide-bar has one particularly useful attribute.  The camera separation can be made small by simply sliding over only a small amount for the second shot.  Thus hypostereo is easy to do.  This is most helpful for closeup photography of a still-life.  Apply the 1/30 rule (for a normal lens).  If the nearest object is 6 inches away, use a separation of about 6/30 = 1/5 of an inch.  This is how I do photos of wall details in slot canyons where the near walls are closer than about 6 feet.  

The wood slide-bar, coupled with an inexpensive slide or print view (see chapter 2),  is a quick way to make some stereo pairs and see if you really like this game.  You will, I think!

DEDICATED 3D CAMERAS

The above paragraphs have shown that it is possible to acquire 3D images with as little equipment as a single film or digital camera.  However, specialized 3D cameras that have the self-contained ability to take two pictures at once have been built and sold in the past,  and a couple of units are being produced today.  Here are some examples:

 

Fig. 1.6 The STEREO REALIST Of commercial film cameras, no company except RBT (see below) is currently manufacturing 3D equipment.  Thus to do good-quality 3D photography using film one can purchase used units and refurbish them using one of several  service companies. The Stereo Realist is a classic from the 1960's.  This camera uses 35mm film and takes pairs of images on a single roll.  The focal length of the lenses is fixed.   Vintage.  Often available used (check Ebay).

 

Fig. 1.7  The BELPASCA Another vintage 35mm camera with high performance lenses.  Somewhat hard to find and sort of a collectors item.

 

Fig. 1.8  SPUTNIK Medium Format Camera A classic Russian-made camera that uses medium format (120 or 220) roll film.  Decent lenses and a plastic body.  Another vintage unit.  Relatively compact and lightweight for generating a 21/4 inch square image.

 

Fig. 1.9  RBT  Model 109 A company, RBT, in Germany, takes standard 35mm cameras and saws the ends off, then recombines them into a stereo unit.  Can have quite good quality lenses, and other advanced (automatic) features.  A camera like this is also quite expensive (about $3000). Ref:  3D Concepts

 

Fig. 1.10  RBT Model S1 Another modern (but no longer in  production) rangefinder type monolithic stereo camera.  Light and compact, this unit is easy to use and carry around.  Price tag again is quite steep, but it produces marvelous images. Ref: 3D Concepts

All the above cameras take pairs of images on a single roll of film.  The pairs line up side by side and the film should be processed (WITHOUT CUTTING OR MOUNTING).  It is necessary to take the film, cut apart the left and right pairs, and mount them in a STANDARD STEREO MOUNT.  The stereo mount is a plastic or cardboard frame that, unlike a slide mount, takes two images and separates them by the typical inter-ocular distance (about 2.4 inches).  Below is a picture of such a mount:

Fig. 1.11  A Stereo Mount After the images are cut out of a 35mm film strip (from one of the above monolithic cameras), they are placed in a frame with holes for the left and right film chips.  The top lid closes over and seals after the desired alignment is achieved.  Both cardboard and plastic frames are available. These mounted pairs are standard for 3D clubs and societies.  They may be viewed with special viewers (see chapter 2: Viewing Methods ), or projected with special projectors (see chapter 7:  3D Projection)

Either (or both) Film and/or Digital cameras can be used for 3D imaging.  For subjects that do not move, a slide bar works well with either type of camera.  For fast-moving subjects, twin-cameras can be used, but need to be closely synchronized.  Further information on this appears in chapter 3 of this Guide and in various Technotes.

In 2003, I found it useful to compare self-contained 3D cameras, such as shown above, with 3D photography using a pair of single cameras.  I now think that for most applications, DIGITAL PHOTOGRAPHY WITH A PAIR OF DIGITAL CAMERAS IS THE BEST METHOD.  Film has limited utility, and makes the whole process of more difficult.   Nonetheless, some might find the comparisons helpful.

ADVANTAGES OF MONOLITHIC 3D CAMERAS:

1. Smaller and lighter than the combination of 2 individual cameras.

2. Shutters generally fire quite close together, sometimes within 1/500 sec  (may be required for action shots).    

3. Flash can often be used (i.e. the shutters are simultaneously fully open when the flash goes off).

4. The mounting format is standard for stereo clubs and societies, WHEN USING FILM.  BUT AS OF NOV. 2004 DIGITAL 3D IS BECOMING MORE POPULAR.  

5. Mounting film clips is somewhat easier than dealing with pairs of 35mm slides.  Film chips may be mounted in frames with registration pins.  These pins allow for horizontal translation and setting the stereo window.

DISADVANTAGES OF MONOLITHIC 3D CAMERAS:

1. The stereo base (lens separation) is fixed.  Hyper or hypostereo is difficult.

2. Lens focal lengths are fixed on most classical units.  Can't easily change angle of view (although some of the high-end RBT cameras come with zooms and have interchangeable lens mounts).

3. Vintage units may have to be repaired and may require more maintenance.

4. The RBT cameras are very expensive - typically 5X the cost of a pair of reasonable 35mm SLR's or digicams, for example.

5. Projection requires special (non 35mm) projectors (either vintage, or quite costly modern units).

ADVANTAGES OF PAIRS (or single) OF 35mm OR DIGITAL CAMERAS:

1. Many people starting in with this already have one or more of these cameras.
2. Stereo base is variable, as are lenses (for SLR's).
3. 35mm film cameras can take full frame images with very high quality optics (most monolithic camera images are smaller than 36mm x 25mm, and most digital cameras do not capture as much detail as a full 35mm frame).

DISADVANTAGES OF PAIRS OF 35mm or DIGITAL CAMERAS:

1. Synchronization of the two shutters, to enable use of a single flash (for example), can be problematic (see more discussion of this in Chapter 3:  Twin-Cams).  This can be a series difficulty, especially for action.
2. Heavier and bulkier (especially when using non-compact SLR's).
3. Slide mounting is generally more difficult.  Any misalignment of the cameras has to be corrected for in the mounting.  Some misalignments cannot be corrected.
4. It may be difficult to mount two cameras with small enough separation for normal 3D.  This is especially bad if you intend to shoot a lot of material with near objects (unless they are static, whence a slide-bar with small lateral translations between pictures can be used).

3D USING DIGITAL CAMERAS

There are some key advantages of digital cameras in the 3D game:

1)  Each shot is really two, so film use is double.  This can get expensive, especially for action shots where you can go through two rolls in a few minutes.   Also there is a new variable in 3D,  the stereo base distance.  Sometimes it is fun to vary this and see what happens.  More film down the tubes.  Digital photography is much more practical costwise.

2)  Digital image registration (in software) is trivial compared with the required mechanical registration of prints and (especially) slides!!!!     This is a key advantage, not to be minimized.

3)  Digital images are easily archived on long-life CD's or DVD's (for example using Verbatim's AZO archival technology).  Digital images are directly printable on good inkjet printers, and are immediately ready for the web.  No slide scanner is needed.

4)  6 megapixel (MP) cameras are now common.  Five megapixel cameras are relatively cheap (costing the same as a year's film supply or less, maybe much less).  At 6MP good 16 x 20 prints are possible, at 2MP good 8 x 10's can be made.  Certainly,  for web based 3D, or the production of stereo card type prints (3" per image), a pair of good-quality 5MP digital cameras works very well for shooting 3D.

5)  Digital cameras can be quite small, which suits them to participatory action photography or other situations where it may be difficult to carry around a 35mm twin SLR rig.

 

There are some minor  disadvantages of digital 3D:

1)  Digital images can be projected, but digital projectors are expensive and the resolution is less, compared with slide projection.   On the other hand, digital projectors are brighter and snappier, and prices are coming down.  Digital projection is the wave of the future.  Many companies (like Kodak) have stopped making slide projectors.   You can of course make slides from your digital images.  One way is just to shoot a high-resolution (2048 x 1556) computer monitor containing the full screen digital images with a film camera .  This actually works pretty well.  

2)  Most self-contained digital cameras won't go to extreme wide angle (i.e. less then 35mm) without adapter lenses (that sometimes distort rather badly).  Stereo often works best with wide angle lenses.   However, for many shots a 35mm lens (35mm equivalent) is OK.

3)  Even at 6MP, digital camera resolution is less than film.  In 3D this is only really important if you will

          a)  Use the hand-held film viewer as you primary way to look at stereos.

          b)   Make prints larger than 16x20 as anaglyphs or for mirror viewing in galleries.

AT THE END OF THE DAY,  DIGITAL IS THE FUTURE OF ALMOST ALL PHOTOGRAPHY.    PLEASE SEE MY "DIGITAL 3D QUICKSTART" ARTICLE.

Technical References (more detail, mathematical equations, graphs.....):

Ferwerda, Jac. G., The World of 3D,  3D Press (Netherlands), 300pp, (the bible) 1990 .

Waack, F.G., Stereo Photography, 78pp, (some graphs and formulae), 1985.

Burder, D.B., Whitehouse, P., Photographing in 3D, Published by Stereoscopic Society of UK. 32pp., 1992 (less technical).

Click here for book sales information.

Return to 3D GUIDE