Greta. A Believable Embodied Conversational Agent

Poggi, Isabella; Pélachaud, Catherine; Rosis, Fiorella de; Carofiglio, Valeria; Carolis, Berardina De

doi:10.1007/1-4020-3051-7_1

Cited by 95 publications

(63 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The captured face animation can be converted to MPEG-4 facial animation parameters (FAPs) [4] for rendering with an avatar e.g. the free MPEG-4 player named GRETA [15].…”

Section: Methodsmentioning

confidence: 99%

Robust 3D Face Tracking on Unknown Users with Dynamical Active Models

Zhou

Horain

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The Active Appearance Models [1] and the derived Active Models (AM) [4] allow to robustly track the face of a single user that was previously learnt, but works poorly with multiple or unknown users. Our research aims at improving the tracking robustness by learning from video databases. In this paper, we study the relation between the face texture and the parameter gradient matrix, and propose a statistical approach to dynamically fit the AM to unknown users by estimating the gradient and update matrices from the face texture. We have implemented this algorithm for real time face tracking and experimentally demonstrate its robustness when tracking multiple or unknown users' faces.

show abstract

“…The captured face animation can be converted to MPEG-4 facial animation parameters (FAPs) [4] for rendering with an avatar e.g. the free MPEG-4 player named GRETA [15].…”

Section: Methodsmentioning

confidence: 99%

Robust 3D Face Tracking on Unknown Users with Dynamical Active Models

Zhou

Horain

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We first use Greta [13], an ECA able to express its emotion through several modalities (verbal and nonverbal modalities). We built a protocol based on the evaluation of two expressive dialogue acts, apologise and rejoice, through a scenario played by the agent.…”

Section: First Experiments On Impactmentioning

confidence: 99%

Artificial companion: Building a impacting relation

Pesty

Duhaut

2011

2011 IEEE International Conference on Robotics and Biomimetics

View full text Add to dashboard Cite

In this paper we show that we are in front of an evolution from traditional human-computer interactions to a kind of intense exchange between the human user and new generation of virtual or real systems -Embodied Conversational Agents (ECAs) or affective robotsbringing the interaction to another level, the "relation level". We call these systems "companions" that is to say systems with which the user wants to build a kind of lifelong relationship.We thus argue that we need to go beyond the concepts acceptability and believability of system to get closer to human and look for "impact" concept. We will see that this problematic is shared between the community of researchers in Embodied Conversational Agents (ECAs) and in affective robotics fields. We put forward a definition of an "impacting relation" that will enable believable interactive ECAs or robots to become believable impacting companions.

show abstract

“…As Poggi et al [1] remark, "Words have formal and informal variants, connotations, positive or negative, tender or insulting nuances". As an exploration of how register is incorporated into the NLG input constructs and the subsequent effect on the generated phrase, normal, formal and informal registers are considered in the current implementation.…”

Section: Multi-context Capabilitiesmentioning

confidence: 99%

“…The NLG component must be able to 'change tone', not only as the context changes in a discourse, but also as the relationship with the user evolves, within and between dialogue sessions -cf. GRETA [1] and the broad research into affect in computing being conducted by the EU network of excellence HUMAINE [2].…”

Section: Introductionmentioning

confidence: 99%

Flexible Natural Language Generation in Multiple Contexts

Cullen

O'Neill

Hanna

2009

Human Language Technology. Challenges of the Information Society

View full text Add to dashboard Cite

Abstract. We present a practical approach to Natural Language Generation (NLG) for spoken dialogue systems. The approach is based on small template fragments (mini-templates). The system's object architecture facilitates generation of phrases across pre-defined business domains and registers, as well as into different languages. The architecture simplifies NLG in well-understood application contexts, while providing the flexibility for a developer and for the system, to vary linguistic output according to dialogue context, including any intended affective impact. Mini-templates are used with a suite of domain term objects, resulting in an NLG system (MINTGEN -MINi-Template GENerator) whose extensibility and ease of maintenance is enhanced by the sparsity of information devoted to individual domains. The system also avoids the need for specialist linguistic competence on the part of the system maintainer.

show abstract

Greta. A Believable Embodied Conversational Agent

Cited by 95 publications

References 20 publications

Robust 3D Face Tracking on Unknown Users with Dynamical Active Models

Robust 3D Face Tracking on Unknown Users with Dynamical Active Models

Artificial companion: Building a impacting relation

Flexible Natural Language Generation in Multiple Contexts

Contact Info

Product

Resources

About