Abstract
Purpose :
Optic flow induced by self-motion can enhance depth perception in 3D space. For observers with reduced acuity and contrast sensitivity, motion cues might be helpful when pictorial and binocular depth cues are unreliable. Motion-related depth cues can be categorized as optical expansion when the observer approaches the fixated object, and motion parallax when the observer’s motion is orthogonal. Is there a difference in the accuracy of depth perception with these two types of observer motion under conditions of blur? If yes, is the benefit due to the difference in starting and ending viewpoints, or does it require continuous motion?
Methods :
We created a virtual 3D room containing a target cube on the floor and a reference sign. Two-second-long videos were rendered in five types: a single static view, optical expansion from continuous forward motion or the corresponding first and last static views, and motion parallax from continuous orthogonal motion or the corresponding first and last static views. Videos were prepared with three levels of calibrated blur—Moderate, Severe, and Profound with corresponding logMAR values of 0.95, 1.15, and 1.55. Ten normally sighted observers viewed the videos on a computer monitor. They judged how far the target cube was in front or behind the reference sign in feet by moving an on-screen slider.
Results :
Averaged across the three blur levels, motion parallax reduced depth estimation errors by 47% and optical expansion by 10% compared to a single static view. The difference in performance between continuous motion and conditions with start-ending views was only significant in the Moderate blur condition (p<.001). In the Severe and Profound blur conditions, the difference between continuous motion and start-and-end static views was not significant.
Conclusions :
Motion parallax yielded greater benefits in judging object depth than did optical expansion. Except in the Moderate blur condition, continuous motion was no more effective than static views at the beginning and end of the motion sequence. Together, these results indicate that motion cues are beneficial in judging object depth for conditions of artificial acuity reduction.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.