Date of Award


Publication Type

Master Thesis

Degree Name



Computer Science

First Advisor

Boufama, B.


Computer Science.



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


Depth recovery is a central problem in Computer Vision. Stereopsis is one of the most extensively investigated techniques amidst the approaches addressing it. Though sufficiently accurate, stereopsis is highly dependent on the success of point correspondence, an ill-posed problem that is known to suffer from a number of underlying difficulties. Single-lens stereo systems emerged recently as an effective means to thwart some of the problems of correspondence. The plenoptic camera in particular, has a special image formation geometry that claims to resolve the correspondence ambiguities and simplify depth recovery. However, to our knowledge, prior to this work, no study has ever been conducted on this system to experimentally assess its viability for depth extraction. The research outlined in this thesis is twofold: the complete simulation process of this camera is presented, which delivers a physically-accurate rendering engine that faithfully accounts for its image formation geometry. Secondly, the depth recovery capabilities of the camera are closely examined; various configurations are tested against different types of scenes using the simulator, and a comparison framework with stereopsis is set up to outline the merits and the disadvantages of this camera. Degrading factors are studied and indicators are given as to how design parameters affect the performance of the system.Dept. of Computer Science. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2003 .A48. Source: Masters Abstracts International, Volume: 42-01, page: 0252. Adviser: Boubakeur Boufama. Thesis (M.Sc.)--University of Windsor (Canada), 2003.