Abstract
Depth perception from motion parallax due to lateral observer translation with fixation uses both retinal image motion and the extra-retinal smooth pursuit signal. When distractor objects are in line-of-sight with the fixate, the relative depth from the fixate is mathematically determined by the ratio of the rate of retinal motion over the tracking rate. When distractor objects are off to the side of line-of-sight, this ratio does not give a good prediction, but the ratio is a time-varying quantity as is the position relative to line-of-sight. In ( Stroyan & Nawrot, 2011, J. Math. Biol.) we showed that the motion/pursuit ratio reaches a maximum value at a position where the fixate and a particular distractor are more, but not exactly, in line. We also showed that the peak ratio can be used to build a point-by-point approximation of the structure of an object.
We show here that the peak ratio for a particular distractor point occurs at the same physical location where the ego-centric relative depth reaches a maximum. This gives a simple explanation of why the peak is not exactly in line-of-sight and why each peak is determined at the same observer position for different translation speeds.
Old experiments (Hildreth, et al, 1990, Perception & Psychophysics, Eby, 1992, Perception & Psychophysics) observed integration times necessary for depth perception an order of magnitude greater than our more recent results. We show how the peak motion/pursuit time could be the cause of this difference for a complicated stimulus shape.
Meeting abstract presented at VSS 2012