Multimedia interaction and access based on emotions:automating video elicited e...

• Author(s): Oliveira, Eva Ferreira de, 1975-
• Date: 2013
• Persistent ID: http://hdl.handle.net/10451/9780
• Origin: Repositório da Universidade de Lisboa
• Subject(s): Filmes - Aspecto cognitivo; Emoções; Psicofisiologia; Computação afectiva; Engenharia informática; Teses de doutoramento - 2013
• Document Type: Doctoral Thesis
• Language: Portuguese
• Advisor(s): Chambel, Maria Teresa Caeiro, 1965-; Ribeiro, Nuno Magalhães, 1969-

Description

Tese de doutoramento, Informática (Engenharia Informática), Universidade de Lisboa, Faculdade de Ciências, 2013 Films are an excellent form of art that exploit our affective, perceptual and intellectual abilities. Technological developments and the trends for media convergence are turning video into a dominant and pervasive medium, and online video is becoming a growing entertainment activity on the web. Alongside, physiological measures are making it possible to study additional ways to identify and use emotions in human-machine interactions, multimedia retrieval and information visualization. The work described in this thesis has two main objectives: to develop an Emotions Recognition and Classification mechanism for video induced emotions; and to enable Emotional Movie Access and Exploration. Regarding the first objective, we explore recognition and classification mechanisms, in order to allow video classification based on emotions, and to identify each user’s emotional states providing different access mechanisms. We aim to provide video classification and indexing based on emotions, felt by the users while watching movies. In what concerns the second objective, we focus on emotional movie access and exploration mechanisms to find ways to access and visualize videos based on their emotional properties and users’ emotions and profiles. In this context, we designed a set of methods to access and watch the movies, both at the level of the whole movie collection, and at the individual movies level. The automatic recognition mechanism developed in this work allows for the detection of physiologic patterns, indeed providing valid individual information about users emotion while they were watching a specific movie; in addition, the user interface representations and exploration mechanisms proposed and evaluated in this thesis, show that more perceptive, satisfactory and useful visual representations influenced positively the exploration of emotional information in movies.