Library · Fundamentals

How stereoscopy works

Our two eyes sit about 6.5 cm apart, and each sees the world from a slightly different place. The brain fuses the two images into one and derives depth from the small differences between them. Stereoscopic 3D cinema (S3D) imitates this arrangement.

Two eyes, two images

Hold a finger in front of your face and close your eyes one at a time: the finger jumps left and right against the background. A distant object barely moves. Near objects sit in very different places in the two eyes; distant ones in almost the same place. The brain perceives this difference as distance. The merging of the two images into one is called fusion.

Top: two eyes and two balls, seen from above. Bottom: the image each eye sees. As the red ball comes closer, its positions in the two images move apart; the distant purple ball is in almost the same place in both images. The differences in the images are drawn four times larger.

In the cinema

In stereoscopic shooting, two cameras take the place of the eyes: one for the left eye, one for the right eye. The device that carries the two cameras together is called a rig, and the two images shot at the same moment are called a stereo pair. The distance between the two cameras, the interaxial, is the main variable that determines how much depth the shot has.

At projection, both images arrive on the same screen, and each eye is made to see only its own image:

Polarized glasses: the two images are projected with different polarization; each lens of the glasses passes only one. Common in movie theaters.
Active glasses: the screen shows the two images in alternation; the lenses of the glasses darken in alternation at the same rate.
Anaglyph: the images are separated by color and viewed with red/cyan glasses. Simple, but it distorts colors.
Head-mounted display: each eye gets its own display.

Depth on the screen

The horizontal difference between the two images on the screen is called parallax. With no difference, the object appears on the screen; with a difference in one direction, it appears behind the screen; in the other direction, in front of the screen. The eyes always focus on the screen but turn toward the distance at which the object appears; this mismatch is the basic difficulty of stereoscopic viewing (see Vergence–accommodation conflict).

The stereographer manages how much depth there is and where it sits by choosing the interaxial and the point the cameras look at. The aim is to keep the depth strong enough to serve the story and restrained enough not to tire the viewer.

Start here

The library is arranged to be read in this order:

Parallax: how the difference on the screen turns into depth.
Convergence: which part of the scene sits on the screen plane.
Interaxial: the amount of depth.
Zero parallax plane: the screen acting as a window.
Then camera rigs, perception, distortions, formulas and post-production.

For an unfamiliar term, see the Glossary; for the letters in the formulas, see the Symbols table.