This project is at the intersection of Visual Perception and Computer Graphics (CG). Computer Graphics finds little in brain science to inform algorithmic depiction of depth because precise quantitative details of the mechanisms that process 2D retinal images into 3D information are poorly known. To advance this challenge this project consists in designing and running psychophysical experiments to calibrate a new CG model for depicting perspective images. Images created using this approach will be shown to human participants who report details of the depth they see in the images. The results of the experiments will improve our knowledge of the role that perspective plays in the quantitative perception of depth.
The goals of the project are to
implement a rendering engine that creates resolution-independent (SVG) 2D images in linear perspective at high resolution,
implement a flexible experimental platform to measure human response to such im- ages using a variety of psychophysical methodologies and
run psychophysical experiments exploring the effect of linear perspective cues on depth perception in 2D images.
Two students will help in the implementation of the web-platform to run the experiments and the rendering engine that generates the stimuli, as well as in the designing, runing and analyzing of the experiments.
Some experience in computer science is required for both positions. For one position knowlege in visual perception and research method in psychology is a strong asset (and one CS course sufficient---COSC 101, COSC 140 or FSEM 131). For the other position the programming experience acquired in COSC 301 (or equivalent) is required.