Rothwell, Sandra (2007) Detection and analysis of emotional highlights in film. PhD thesis, Dublin City University.
Abstract
Thls work explores the emotional experience of viewing a film or movie. We seek to invesagate the supposition that emotional highlights in feature films can be detected through analysis of viewers' involuntary physiological reactions. We employ an empirical approach to the investigation of this hypothesis, which we will discuss in detail, culminating in a detailed analysis of the results of the experimentation carried out during the course of this work. An experiment, known as the CDVPlex, was conducted in order to compile a ground-truth of human subject responses to stimuli in film. This was achieved by monitoring and recordng physiological reactions of people as they viewed a large selection of films in a controlled cinema-like environment using a range of biometric measurement devices, both wearable and integrated. In order to obtain a ground truth of the emotions actually present in a film, a selection of the films used in the CDVPlex were manually annotated for a defined set of emotions. We examine how filmmakers use devices and techniques to stimulate viewers' emotions, particularly how music is used in film to intensify the impact of onscreen action. We also examine the different event types of which film scenes are comprised and how they can be detected using audio-visual analysis of the video content. Finally, we calculate and study the correlations between the emotions found in the annotated films, and "events" or highhghts in the viewers' biometric measurements, and also between the scene events and music occurring in the film and the emotions present.
Metadata
Item Type: | Thesis (PhD) |
---|---|
Date of Award: | 2007 |
Refereed: | No |
Supervisor(s): | Smeaton, Alan F. |
Uncontrolled Keywords: | emotional reactions; physiological reactions; bimoetric measurements |
Subjects: | Humanities > Film studies |
DCU Faculties and Centres: | DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing |
Use License: | This item is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 3.0 License. View License |
ID Code: | 17012 |
Deposited On: | 15 May 2012 13:22 by Fran Callaghan . Last Modified 19 Jul 2018 14:55 |
Documents
Full text available as:
Preview |
PDF
- Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
11MB |
Downloads
Downloads
Downloads per month over past year
Archive Staff Only: edit this record