Virtual characters express their emotions
Spanish researchers have developed a computer model able to generate virtual faces which display emotions and moods according to personality traits. The team leader explains: 'The aim of this work has been to design a model that reveals a person's moods and displays them on a virtual face. In the same 3-D space we have integrated personality, emotions and moods, which had previously been dealt with separately.' This model could be applied in both educational environments (virtual tutors and presenters with personality traits) and in video game characters. But read more...
You can see above a virtual character showing various universal emotions: anger, disgust, fear, joy, sadness and surprise. (Credit: Diana Arellano, UIB). Here is a link to a larger version of this picture. You'll find more facial expressions on the research project website, Creación y Animación de Expresiones Faciales where you'll be greeted by Alice, the natural language processing chatterbot (Credit: Wikipedia). If Alice speaks only in English, the website is written in Spanish. It is focused on AFE (Affective Facial Expressions) or the creation of facial expressions to simulate emotions and personality traits.
This research project has been conducted at the Computer Graphics and Vision Group (UIB), a research unit of the Mathematics and Computer Science Department (site in Spanish) at the University of Balearic Islands. This research project has been led by Diana Arellano, a PhD student, under the supervision of two UIB professors, Xavier Varona and Francisco J. Perales.
What were the scientific basis for this project? "The researcher pointed out that emotions (such as fear, joy or surprise) are almost instantaneous mood alterations, in contrast to emotional states (such as boredom or anxiety) which are more long-lasting, or personality, which normally lasts someone's entire life. The designers have followed the theories of Albert Mehrabian [Credit: Wikipedia] to draw up the model, based on the five personality traits established by this American psychologist: extraversion, neuroticism, openness, conscientiousness and agreeableness. 'Every personality can be considered an emotional state by default,' indicated Arellano. An introverted and neurotic personality is therefore related to an anxious emotional state. The points of the face that define these emotions can be determined mathematically, and the algorithms developed by computer experts can be used to obtain different facial expressions 'quickly and easily.'"
And is this technique successful? Apparently yes. "The results of the method have been assessed objectively (through an automatic recognizer which identified 82% of the expressions generated) and subjectively, through a survey carried out among a group of 75 university students. The students successfully recognised 86% of the emotions and 73% of the emotional states shown on the computer."
This research work has been presented at the Computer Animation and Social Agents 2008 Conference (CASA 2008) which was held on September 1-3, 2008 in Seoul, South Korea. CASA 2008 was also the focus of a special issue of the Computer Animation and Virtual Worlds magazine published by Wiley InterScience.
This work has been published by the magazine on July 25, 2008 under the title "Generation and visualization of emotional states in virtual characters" (Volume 19, Issue 3-4, Pages 259-270). Here is a link to the abstract. "This paper presents an affective model that determines the emotional state of a character according to the personality traits and the experienced emotions. We consider an emotional state as the layer between personality and emotion. The proposed affective model offers a mapping between emotions and emotional states. To evidence emotional states of a virtual character, we can attribute them facial expressions based on their associated emotions. Facial expressions for intermediate emotions are generated automatically from expressions for universal emotions. The experiments show coherent emotional states produced by a simulated story. They also present how the corresponding emotions were represented through dynamic and static facial expressions. Finally, the obtained results demonstrate the satisfactory recognition by a group of people unfamiliar with the work described."
Sources: Scientific Information and News Service (Servicio de Información y Noticias Científicas - SINC) news release, December 4, 2008; and various websites
You'll find related stories by following the links below.