Faculty Profiles - ITOU Junko

Misc - ITOU Junko

Division display221 - 233 of about 233／ All the affair displays >>

Proposing the Interdependencies Model between Nonverbal Conversational Signals Displayed by Embodied Agents in a Dialogue

ITOU Junko, KAKUSHO Koh, MINOH Michihiko

IEICE technical report ( The Institute of Electronics, Information and Communication Engineers ) 105 ( 433 ) 13 - 18 2005.11

This paper proposes a method of properly controlling nonverbal conversation signals(NCS) displayed by embodied agents which inform users through their dialogues, according to the atmospheres. The controls is important not to convey misunderstanding meanings from their NCS to users. It reports in the field of social psychology that there are some interdependences between NCS expressed in person-to-person dialogue. We focus on the interdependences and classify each scenes of dialogue, aiming to inform audiences, into three essential situations. Then, we define the three situations as atmospheres, and propose a procedure to maintain the interdependences between NCS, which vary with the atmospheres. Finally, we experiment using NCS extracted by TV programs aiming to inform users through dialogues between TV personalities, and evaluate the NCS realized by our procedure.

Learning Mapping of Facial Expressions between User and Lifelike Agent in the User's Subjective View for Agent-Mediated Facial Communication

KAKUSHO Koh, LI Liqun, ITOU Junko, MINOH Michihiko

Journal of Japan Society for Fuzzy Theory and Intelligent Informatics ( Japan Society for Fuzzy Theory and Intelligent Informatics ) 17 ( 3 ) 325 - 339 2005

In order to transmit facial expressions of the user using the face of a lifelike agent in agent-mediated distance communication, we have to realize facial expression mapping, which gives synthetic facial expressions of the agent appropriate for representing facial expressions of the user in communication. However, the criterion to determine which expression on the agent's face is the "same" as that on the user's face heavily depends on the subjective view of the user. This article discusses how to learn the facial expression mapping reflecting the user's subjective view from the interaction with the user for agent-mediated communication. In the discussion, we focus on the two problems: how to make the facial expression mapping realized by the system reflect the user's subjective view gradually through the interaction, and how to realize the facial expression mapping with the error less than the range acceptable for the user. We propose an approach that copes with these two problems by (1) acquiring positive examples and negative ones for the facial expression mapping reflecting the user's subjective view from the interaction with the user, (2) estimating the acceptable range of the error to realize the facial expression mapping, and (3) constituting RBF(Radial Basis Function) network that realizes the facial expression mapping within the estimated range of the error.

DOI

Modeling the Interdependency between Nonverbal Conversational Signals for Conversational Agents

KAKUSHO Koh, ITOU Junko, MINOH Michihiko

IPSJ SIG Notes. ICS ( Information Processing Society of Japan (IPSJ) ) 134 31 - 38 2003.10

In this article, we discuss how to control nonverbal conversational signals (NCS) including eye gazes, noddings and facial expressions displayed by a pair of embodied agents for information presentation using their dialogues. Recent work on information agent proposes to present information to the users in the form of dialogues between a pair of embodied agents, aiming at clarifying the point that the users should be interested in. However, when agents have faces and bodies in order to be "embodied", the users read various meanings in the NCS displayed by the faces and the bodies of the agents, even if the agents are not actually designed to send NCS to their users but only to speak. Thus, we need to control the NCS of the embodied agents properly so that they convey appropriate meaning to the users. In order to cope with this problem, there is some work that discusses how to control NCS of an embodied agent. That work mainly focuses on the consistency of NCS with the speech utterances or the task assumed for the agent. However, when we employ dialogues between a pair of embodied agents for information presentation, we need to consider interdependences between NCS displayed by those two agents. In this article, we will propose a method for adjusting NCS displayed by a pair of embodied agents so that the interdependences between NCS are maintained. We will first explain the knowledge about the interdependences between NCS in human communication reported in the field of social psychology. Then, we will propose a procedure to maintain the interdependences between NCS of embodied agents in dialogues by satisfying constraints imposed on the NCS. Experimental results for evaluating the NCS realized by our procedure will also be given at the end of the article.

Adjustment of Nonverbal Conversational Signals from Embodied Agents in Dialogues

Junko ITOU, Koh KAKUSHO, Michihiko MINOH

7th International Conference on Knowledge-based Intelligent Information & Engineering Systems(KES2003) 2003.09 [Refereed]

Control of Atmosphere Based on the Intimacy Equilibrium Model During the Dialogue Between Computer Graphics Animated Characters

ITOU Junko, KAKUSHO Koh, MINOH Michihiko

Technical report of IEICE. HIP ( The Institute of Electronics, Information and Communication Engineers ) 102 ( 735 ) 37 - 42 2003.03

We discuss a problem of control the atmosphere during the dialogue between computer graphics (CG) animated characters. The control is maintained by the adjustment of their visual nonverbal expressions. There are some results on research incorporating translation of text based information to spoken dialogue between CG characters. We focus on the impact of visual nonverbal expressions of the CG characters on the communication atmosphere. In the real world a lot of interdependencies of nonverbal expressions during the person-to-person dialogue may be observed. The intimacy equilibrium model, that is one of the strategies among research communities to model such interdependencies, describes each interpersonal relationship with the assumption of the balance in the intimacy presented using their nonverbal expressions. Based on that balance, we propose a new model to control the atmosphere by adjusting the character's nonverbal expressions adequately.

親密性平衡モデルに基づくCGキャラクタ同士の会話映像における雰囲気の制御

伊藤淳子, 角所考, 美濃導彦

シンポジウム報告集 ( 京都大学学術情報メディアセンター研究開発部 ) ( 1 ) 61 - 70 2003.03

擬人化エージェント同士の対話における視覚的ノンバーバル表現の制御

伊藤淳子, 角所考, 美濃導彦

人工知能学会全国大会論文集 ( 一般社団法人人工知能学会 ) 3 ( 0 ) 127 - 127 2003

DOI

An embodied agent that sends nonverbal conversational signals consistent with those of the partner during a dialogue

Koh Kakusho, Junko Itou, Michihiko Minoh

Proceedings - IEEE International Workshop on Robot and Human Interactive Communication 247 - 252 2003 [Refereed]

In this article, we discuss how to control nonverbal conversational signals (NCS) including eye gazes, nods and facial expressions displayed by a pair of embodied agents in dialogue. Recent work on information agent proposes to employ embodied agents for presenting information to the viewers. Since embodied agents have faces and bodies in order to be "embodied", the viewers read various meanings in the NCS on the faces and the bodies of the agents, even if the agents are not actually designed to send NCS to their viewers but only to speak. In order to cope with this problem, it has been proposed in some previous work to make the NCS appropriate for the speech utterances or the task of the agent. However, when two embodied agents are in dialogue, we also need to consider interdependences between NCS displayed by those agents. We discuss how to maintain those interdependencies for producing dialogues between a pair of embodied agents for information presentation. © 2003 IEEE.

DOI

K-14 Understanding Atmosphere of Communication from Observation of Human Faces

Itou Junko, Kakusho Koh, Minoh Michihiko

情報科学技術フォーラム一般講演論文集 ( Forum on Information Technology ) 2002 ( 3 ) 395 - 396 2002.09

Observing Human Faces in Communication to Understand the Atmosphere

Junko ITOU, Koh KAKUSHO, Michihiko MINOH

International Workshop on Intelligent Media Technology for Communicative Reality in conjunction with PRICAI-02 2002.08 [Refereed]

顔メディアに注目したコミュニケーションの分析

伊藤淳子, 角所考, 美濃導彦

第64回全国大会講演論文集 2002 ( 1 ) 555 - 562 2002.03

A solution of lighting problems in Image-Based Rendering

ITOU Junko, SHEN Li, BABAGUCHI Noboru, KITAHASHI Tadahiro

Technical report of IEICE. PRMU ( The Institute of Electronics, Information and Communication Engineers ) 100 ( 702 ) 71 - 78 2001.03

Usual Image-Based Rendering(IBR) has disadvantage in creating an image under an arbitrary illumination environment. Several peoples have discussed this illumination problem on the basis of picture sequences taken in many different illumination environments at a fixed point, based on the theory of the dichromatic model. They take heuristic approaches to deal with the specular reflection component, while some present a solution to the diffuse reflection component using the theory. In this paper, we present a procedure to create an image of dielectric objects from a virtual camera set in an arbitrary environment at an arbitrary position. The image is obtained from three image sequences taken in three kinds of illumination environments. The procedure includes a novel proposal of determining the points of specular refletion on the objects by calculating them using the theory of reflectance map.

文書画像に対するHilbert 走査の応用可能性について

伊藤淳子, 北橋忠宏

第61回全国大会講演論文集 2000 ( 1 ) 245 - 246 2000.10

　 PREV -NEXT　

<ITOU Junko>

Misc - ITOU Junko