Visibility and acceptance of continuous parallax distortions
A few words
Our project revolves around 2 basic axes; the first lies in understanding the human visual component and the underlying mechanisms in processing the information from a finitely reconstructed version of reality and the second is focused on how we can exploit some known limitations of the human visual system, initially to improve the visual experience of continuous parallax content by also quantifying this subjective factor, and subsequently to possibly direct research to areas of greater importance or help in improving current compression schemes.
As an initial step we mainly examined effects related to motion parallax and binocular vision by analyzing the artifacts which arise in image sequences simulating the changing views of a 3-D object as the viewer moves around it. Our goal was to specify the minimal step in changing the views density which is noticeable by investigating when the related artifacts become visible. We attempt to predict this visibility by using a mathematical model based on the spectral content of the image sequences and to define a perceptual threshold. We have simulated these effects on a stereo display and performed psychophysical studies on human participants to determine our model validity.
Christos has a diploma in applied mathematical and physical sciences from the National Technical University of Athens, Greece, and a Master’s degree in computational mechanics from the same institution. He has worked on human perception and kinematics in three dimensional space with Prof. K Siettos and Prof. N Smyrnis at the Laboratory of Sensorimotor Control of the Mental Health Research Institute in Athens.
Henry Wellcome Building, Institute of Neuroscience
Newcastle University Medical School, NE2 4HH, UK