T3D optics of stereo photography

[Peter Abrahams <telscope@xxxxxxxxxx>]

This is an essay I wrote for the big Cascade Stereo Club annual show,
2/28-3/1 in Portland.

The Optics of Stereo Photography   by Peter Abrahams
In a person’s visual field are seen line, shape, size, color, brightness or
darkness, and other qualities.  For people with two normally functioning
eyes, there are many other aspects of vision that provide information on
the distance of objects from the observer.  These cues for depth are of two
types.  One can be seen with one eye and the other requires the use of both
eyes to be seen.
Monocular cues include the following.  The relative sizes of objects, since
large objects seem nearer than small ones.  An object that blocks the view
of another object appears closer than the other.  Distant objects appear
dimmer and have muted colors.  Movement of the observer causes the near
objects to shift relative to objects in the distance.  The eye must make an
effort to focus on near objects, and while this might not be felt, it does
add to what the brain knows about the objects.
Binocular cues to the distance of objects from the observer use the
coordinated function of both eyes to quickly & precisely give a sense of
depth.  For example, when viewing a close object, the eyes use their
muscles to rotate inward and increase their focus, and the sensation from
the ocular muscles gives a cue to the brain that the object is nearby.
However, the most important cue to depth is stereoscopic vision, the basis
for stereo photography and most three dimensional imaging.  The two eyes
are separated by about 65 mm, and each eye sees a slightly different view
of the object.  These two images are combined in the brain, and the
differences between the two allow the brain to reconstruct a model of the
object that includes the dimension of depth.
Stereo photography uses a camera with two lenses, or two separate cameras,
to produce a pair of images that are similar to what the two eyes of an
observer would see.  If the left photo is presented to the left eye and the
right photo to the right eye, the observer can fuse them into a 3d image,
just like the images of the real world are fused.  This can allow
photography of great beauty, and also very useful & functional images, used
by scientists, designers, teachers, and others.
The two cameras have lenses that are separated by 2 or 3 inches, and the
images on the two negatives are slightly different.  An object that is
straight in front of the cameras will be registered on the left film at a
location slightly to the left of where it appears on the right film.  This
same outward shift occurs on the right negative.  When the film is exposed,
the images are upside down in the frame.  When mounted as slides, they are
rotated, so the image is erect in the frame.  This reverses left and right,
and causes the outward shifting of the images to ‘become’ inward shifting.
When the two frames are viewed in a viewer (assuming slide film, the same
principles apply to prints), the eyes will have to converge slightly to see
both images.  This is identical to the convergence of the eyes towards an
object directly in front of the face.  The two lenses registered slightly
different views of the object, which the brain fuses into one
three-dimensional image.
To replicate an observer’s view of an object, the two camera lenses are
separated by a distance equal to the typical distance between the centers
of the eyes (the interpupillary distance).  Then the resulting stereo
photograph is viewed from a certain distance, so that the objects in the
picture occupy the same area in the visual field as the real objects did.
To do this, the lenses of the viewer will be the same focal length as the
lenses of the camera.  This allows the brain to reconstruct the object as
an image that appears to have the same dimensions and relation to the space
around it, that the object has.  If stereo photography is used to document
or record a scene, these standards should be adhered to.  However,
stereography is capable of much more than just documentation.  The third
dimension of depth can be manipulated in many ways, although a very
conservative technique is always recommended, as extremes can cause
headaches and eyestrain.
By placing the two cameras farther apart, the dimension of depth in the
photos is exaggerated.  The increased distance between cameras will cause
near objects to be shifted relative to far objects in the photos.  This
‘disparity’ is interpreted by the brain as depth.  Stereo photos of tiny
objects will show similar increased depth unless the camera lenses are
placed closer together.
Other artistic effects can be obtained when mounting the slides or photos.
If they are spaced close together, so that the eyes to converge on the
subject of the photo, the object can appear to be a tiny model at a close
distance, because we normally converge the eyes to see tiny objects, and
the brain interprets the convergence as ‘tiny object in view’.  When
projected onto a screen, the object will appear to be floating in front of
the screen (an undesirable effect).
If the slides are mounted farther apart in their holder, the eyes will look
straight ahead, with no convergence, to see the pair of subjects.  The
subject can appear to be a much larger object at a farther distance from
the viewer, because we normally look straight ahead to view at a distance.
When projected, the object will appear at a distance behind the screen.
The ideal for projection is for the object to appear just behind the screen.

Greenleaf, Allen.  Photographic Optics. N.Y.: Macmillan, 1950.
Kingslake, Rudolf.  Optics in Photography.  Bellingham: SPIE, 1992.
Ray, Sidney. Applied Photographic Optics. Oxford: Focal Press, 1994.

_______________________________________
Peter Abrahams   telscope@xxxxxxxxxx
the history of the telescope, the microscope,
    and the prism binocular


------------------------------