Purchase this article with an account.
David Tinjust, Remy Allard, Jocelyn Faubert; Impact of stereoscopic vision and 3D representation of visual space on multiple object tracking performance. Journal of Vision 2008;8(6):509. doi: https://doi.org/10.1167/8.6.509.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Classical multiple object tracking (MOT) studies use 2D visual space representation. Also, they do not take into account stereoscopic vision capacity that allows better discrimination between the relative positions of multiple objects in space. However, our reality is a 3D world where multiple objects move at different depth position with different speeds. We have conducted several experiments to evaluate the impact of different non-stereoscopic and stereoscopic representation of space on MOT performance. Moreover, instead of measuring the number of targets that can be tracked, we have used a new kind of measure based on the evaluation of the greatest speed at which the observer is capable to track a set of moving targets (four targets). This kind of measure allows a more precise threshold measurement to discriminate the performance of two observers that can track the same number of targets. The results of our experiments have shown that, relative to the non-stereoscopic conditions, significantly better speed thresholds were obtained with the stereoscopic representations of space. These results suggest that to better conform to our reality, 3D representation of the visual space should be use to optimally measure MOT performance. Finally, contrary to the classical method using the number of objects tracked, the evaluation of the speed threshold for a set of moving target appears to be a better representative measure to differentiate MOT performance between individuals.
This PDF is available to Subscribers Only