Library · Formulas
Orthostereoscopy
Reproducing the scene at its real sizes and distances. The cameras are placed where the viewer's eyes are. The first systematic theory from shooting to projection was set out by Spottiswoode and Spottiswoode (1953).
b = e C = V (for full scale, also f / Wc = V / Ws)
Devernay and Beardsley give this condition in general form:
Wconv / Ws = C / V = b / e
where: Wconv the width covered by the frame at the convergence distance, Ws screen width
If the three ratios are equal, the scene is perceived at one-to-one scale with linear depth: a point at distance Z from the camera appears at distance Z from the viewer.
Limits
- Valid only for a single viewing condition. When the screen or the viewing distance changes, the scene scale and the depth ratios are distorted.
- Full scale ties the angle of view of the lens to the viewing angle; no freedom of composition is left.
- Because convergence is placed at the viewing distance, the part of the scene nearer than this distance comes out in front of the screen.
- At full scale the parallax of a point at infinity is exactly the interocular distance (P∞, the screen parallax of a point at infinity, equals e); the back limit is the divergence limit itself.
The theory is the reference point against which the other methods are measured: depth consistency aims to keep the orthostereoscopic ratio around the main subject, that is, to hold roundness at 1.