Generating Virtual Characters from Personality Traits via Reverse Correlation and Linear Programming (original) (raw)

Exploiting Reverse Correlation for the Generation of Virtual Characters from Personality Traits

Proceedings of the 7th International Conference on Intelligent Technologies for Interactive Entertainment, 2015

Judging from appearance, most people assign personality traits to virtual characters. This paper presents a platform capable of generating online Reverse Correlation experiments for studying the relations between the appearance of virtual characters and their assumed personality. The method used by the platform, which leverages crowdsourcing and interactive genetic algorithms, can be used to generate virtual characters starting from a description of their personality. The platform requires a training phase to gather the beliefs and convictions that people have when judging a person from his/her appearance. The method is validated through two experiments. The first experiment provides evidence on how effective the method is in improving the mapping through genetic evolution. The second experiment illustrates how the method relates to the technique of Reverse Correlation to infer which physical attributes contribute to the perception of a specific trait.

Personality models to and from virtual characters

2017

In order to be believable, virtual agents must possess both a behavioral model simulating emotions and personality, and a convincing aesthetics [4]. A lot of research already exists about models of emotions, and some seminal work investigates now the role of personality [1, 2]. While emotions are dynamic and variable in time, personality is a static feature of humans, changing only very slowly through the course of a life. The emotional state drives the style of the behavior of a character (how it accomplishes actions) the personality drives the intention of an autonomous agent (what to do next). However, there is not much work investigating the relationships between the personality of a virtual agent, its behavior, and its physical appearance. The work that we are conducting in the SLSI group is based on the observation that people very quickly build up their ideas about the personality of others in zero-acquaintance encounters [11]. The judgment of the personality can be modeled, ...

Employing personality-rich virtual persons - New tools required

Computers & Graphics, 2007

Digital personalities are virtual characters that possess situation dependent, personality revealing behaviors. This paper introduces concepts, current and prospective usages of personality-rich virtual characters, and focus on authoring issues for artists that want to create digital portrays of digital persons. Different categories of digital personalities are introduced, in particular virtual actors for CGmovies are explained in more detail because of the importance of understanding the functioning of virtual actors-other types of virtual personalities will often be build upon the faculties of virtual actors. It is proposed that an essential feature of the prospective authoring process of digital personalities is the support of generalization. This would allow for a ''concrete authoring'', that modifies and defines concrete cases. Thus, the author would be able to employ concrete cases as basis for generalizations. In this context, possible technologies that would support a concrete authoring process are presented, in particular Case-Based Reasoning. r

Imparting individuality to virtual humans

2002

In this paper, we present an integrated method of linking personality and emotion with the response generation and expression synthesis of Virtual Humans. The characters are powered by a dialogue system that consists of a large set of basic interactions between user and computer. These interactions are encoded in finite state machines. Transitions are linked with conditions and actions that can be connected with external modules. One of these modules is a personality module. In this way, responses of the virtual human depend not only on input given by a user, but also on its personality and emotional state. The dialogue system is connected to a 3D face that performs the speech and facial animation, together with facial expressions that reflect the personality specification.

Modeling Warmth and Competence in Virtual Characters

Lecture Notes in Computer Science, 2015

Developing believable virtual characters has been a subject of research in many fields including graphics, animations, artificial intelligence, and humancomputer interaction. One challenge towards commoditizing the use of virtual humans is the ability to algorithmically construct characters of different stereotypes. In this paper, we present our efforts in designing virtual characters that can exhibit non-verbal behaviors to reflect varying degrees of warmth and competence, two personality traits shown to underlie social judgments and form stereotypical perception. To embark on developing a computational behavior model that portrays these traits, we adopt an iterative design methodology tuning the design using theory from theatre, animation and psychology, expert reviews, user testing and feedback. Using this process we were able to construct a set of virtual characters that portray variations of warmth and competence through combination of gestures, use of space, and gaze behaviors. In this paper we discuss the design methodology, the resultant system, and initial experiment results showing the promise of the model.

Varying Personality in Spoken Dialogue with a Virtual Human

Lecture Notes in Computer Science, 2009

We extend a virtual human architecture that has been used to build tactical questioning characters with a parameterizable personality model, allowing characters to be designed with different personalities, allowing a richer set of possible user interactions in a training environment. Two experiments were carried out to evaluate the framework. In the first, it was determined that personality models do have an impact on user perception of several aspects of the personality of the character. In the second, a model of assertiveness was evaluated and found to have a small but significant impact on the users who interacted with the full virtual human, and larger differences in judgement of annotators who examined only the verbal transcripts of the interaction.