Presentation 2011-08-26
Estimating User's Conversational Engagement based on Head Pose Information
Ryota Ooko, Ryo Ishii, Yukiko Nakano,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) With the goal of building an intelligent conversational agent that can recognize user's engagement, this paper proposed a method of judging the user's conversational engagement based on the head pose data. First, we analyzed how head pose information is correlated with user's conversational engagement, and found that amplitude of head movement and rotation has positive moderate correlation with the level of conversational engagement. Then, we established an engagement estimation model by applying a decision tree learning algorithm to 19 parameters. As a result, it was revealed that the proposed model based on the head pose information performs quite well.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) conversational engagement / head pose / eye gaze
Paper # HCS2011-32
Date of Issue

Conference Information
Committee HCS
Conference Date 2011/8/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Human Communication Science (HCS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Estimating User's Conversational Engagement based on Head Pose Information
Sub Title (in English)
Keyword(1) conversational engagement
Keyword(2) head pose
Keyword(3) eye gaze
1st Author's Name Ryota Ooko
1st Author's Affiliation Graduate School of Science and Technology, Seikei University()
2nd Author's Name Ryo Ishii
2nd Author's Affiliation NTT Cyber Space Laboratories, NTT Corporation:Graduate School of Informatics, Kyoto University
3rd Author's Name Yukiko Nakano
3rd Author's Affiliation Dept. of Computer and Information Science, Seikei University
Date 2011-08-26
Paper # HCS2011-32
Volume (vol) vol.111
Number (no) 190
Page pp.pp.-
#Pages 5
Date of Issue