Depth perception in stereoscopic displays

Abstract

Abstract— Across two experiments, six factors that could affect depth perception in stereoscopic displays were investigated. In both experiments, the perceived depth of a square‐shaped target was measured and compared to predictions derived from geometry. In Experiment 1, the effects of half‐image separation magnitude, separation direction (crossed vs. uncrossed), viewing distance, stimulus size, and exposure duration were examined. Increases in separation and distance produced increases in depth consistent with predictions. There was a tendency for crossed depth to be slightly less, and uncrossed depth to be greater, than predictions. In Experiment 2, the effect of half‐image onset asynchrony (delaying onset of one eye's image relative to onset of partner eye's image) was investigated. Perceived depth was consistent with predictions under asynchrony conditions, except for very brief exposures of uncrossed depth where reliability of depth perception was degraded. Implications of these results for the design of stereoscopic displays are discussed.