Use this URL to cite or link to this record in EThOS:
Title: Lecture adaptation for students with disabilities
Author: Hughes, G. F.
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 2009
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
In this dissertation, I introduce new methods of adapting lectures for students with disabilities to facilitate their learning in a higher education setting. These new methods use off-the-shelf computer hardware and specialised algorithms to address the specific needs of students with disabilities in a lecture environment. The techniques are able to overcome the problems of traditional adaptation techniques while also providing benefits that were previously unavailable to students with disabilities. This dissertation introduces the Photonote system, which was created to validate the effectiveness of my new adaptation techniques. The system can capture video of the lecturer, and sign-language interpreter if applicable, through the use of standard-definition video camera. Photonote uses a high-resolution digital-still camera to capture presented visual information in a classroom. Captured images are processed by my algorithm to compensate for angular distortion, remove obstructions, and extract a high-contrast representation of the presented information. This approach provides a clear copy of information presented in a lecture, tailored to meet the needs of a user with a disability. Most importantly, the Photonote system is able to capture, process, and display visual information presented to an individual without requiring a lecturer to change his or her presentation technique. This system was evaluated through a large user study that determined the system’s effectiveness at improving on existing classroom-adaptation methods.  The evaluation demonstrated that test scores for students with disabilities can be improved through the use of the new adaptation techniques. This dissertation also investigates methods of automatically determining the most relevant piece of presented visual information at a given point in time during a lecture. A new approach is presented that uses the head pose of audience members to infer the location of the most relevant visual information. A new head-pose detection method and algorithm are discussed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available