Animating Conversation in Online Games

Abstract

Abstract. When players in online games engage in conversation with each other, often through a chat window, their graphical avatars typically do not exhibit interesting behavior. This paper proposes using a model of face-to-face communication to automatically generate realistic nonverbal behavior in the avatars based on what is going on in the conversation. It describes Spark, a flexible XML based architecture that makes this possible and reports on a user study that shows how such avatars could significantly improve online conversations.

Spontaneous Avatar Behavior for Human Territoriality

“…In fact, they reported they felt even more in control, suggesting that the automated avatars were providing some level of support (Cassell & Vilhjalmsson 1999). The Spark system took this approach further by incorporating the BEAT engine (Cassell, Vilhjalmsson & Bickmore 2001) to automate a range of discourse related co-verbal cues in addition to cues for multi-party interaction management, and was able to demonstrate significant benefits over standard chat interaction in online group collaboration (Vilhjalmsson 2004). Focused more on postural shifts, the Demeanor system (Gillies & Ballin 2004) blends user control at several different levels of specification with autonomous reactive behavior to generate avatar posture based on affinity between conversation partners.…”

Social Puppets: Towards Modular Social Animation for Agents and Avatars

“…Such tools both exist for the more abstract sociopsychological models [16], [18] and for the rule-based generation of nonverbal behavior [5], [10]. The work presented here on Social Puppets, a special tool for game environments, is very much influenced by the latter, with roots in the Spark framework for animating online avatars [22] and its core engine which itself was based on the BEAT nonverbal behavior toolkit [5]. The Social Puppets approach aims to accommodate any kind of higher level agent models and lower level animation systems by supplying a clear behavior interface.…”

“…For example, the manager in Tactical Pashto implements the turn-taking model from [22] where it is assumed that the turn is returned to whomever spoke before the current speaker if no explicit turn action is taken. Because the manager keeps track of all puppets and their groups, it is a good place for implementing top-down behavior models for group behavior, whereas the puppets themselves are a better place for bottom-up rules that are meant to result in some emergent social order.…”

Applying Direct Manipulation Interfaces to Customizing Player Character Behaviour

“…The two types of behaviour we are using are gesture which has been studied by Cassell et al [4] and posture which has been studied by Cassell et al [5] and by Bécheiraz and Thalmann [2]. Vihljàlmsson [28] has applied this type of autonomous non-verbal behaviour to avatars for on-line games.…”