Depth-of-field in (semi-)natural photographs

Abstract

Depth of Field is roughly defined as the depth range that is perceived to be sharp in the photograph. Although earlier research has shown effects of depth of field in simple scenes, it is unclear what the effect of depth of field is in semi-natural scenes in which there is a redundancy of depth cues. I made four different semi-natural scenes consisting of a central and two flanking figurines. For each scene, I made a series of photos in which the depth between the central the flanking figurines was increasing in small steps. I made these photo series with six different Depths of Field. In the first experiment participants had to adjust the position in depth of the two flanking figurines relative to a central figurine, by toggling through the photos, such that the perceived depth equaled the perceived lateral distance between the two flanking figurines. Viewing was either monocular or bi-ocular (non-stereo). In the second experiment, I varied the viewing distance. I found that participants' depth / width settings increased with increasing Depth of Field. In other words, when the Depth of Field increased, the perceived depth decreased. Participants made lower settings for monocularly seen stimuli than for bi-ocularly seen stimuli and decreased when the viewing distance was increased. The effect sizes of viewing mode and viewing distance were in the same order as the effect size of the Depth of Field over the experimental range. I conclude that in photographs of (semi-) natural scenes, in which depth cues within the image are highly redundant, and where there are plenty of conflicting depth cues as well, depth of field is an important cue that is taken into account in the perception of depth.