Presentation 2003/11/14
Head Gesture Recognition for the Moving Conversation Robot
Yasushi EJIRI, Kei NAKAJIMA, Shinya FUJIE, Yosuke MATSUSAKA, Tetsunori KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A novel head gesture recognition system for conversation robots is realized, which works well even with distorted input images caused by the movements of the robots. It is very important for the conversational robot to recognize head gestures of conversational partner, because they play important role to realize smooth communication. In this study, we firstly investigate a base system of gesture spotting, which uses HMM and optical flow-based feature extraction. This base system works well when the robot is inactive. However, the performance is degraded when the robot itself is performing gestures since the input images are seriously distorted in this case. To solve this problem, a compensation method of feature parameters using optical flow of body area is proposed, and the MLLR model adaptation technique is adopted. Experimental results show the effectiveness of these techniques.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) head gesture recognition / optical flow / HMM / MLLR / conversation robot
Paper # PRMU2003-153,HIP2003-59
Date of Issue

Conference Information
Committee HIP
Conference Date 2003/11/14(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 Information Processing (HIP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Head Gesture Recognition for the Moving Conversation Robot
Sub Title (in English)
Keyword(1) head gesture recognition
Keyword(2) optical flow
Keyword(3) HMM
Keyword(4) MLLR
Keyword(5) conversation robot
1st Author's Name Yasushi EJIRI
1st Author's Affiliation Department of Computer Science, Waseda University()
2nd Author's Name Kei NAKAJIMA
2nd Author's Affiliation Department of Computer Science, Waseda University
3rd Author's Name Shinya FUJIE
3rd Author's Affiliation Department of Computer Science, Waseda University
4th Author's Name Yosuke MATSUSAKA
4th Author's Affiliation Department of Computer Science, Waseda University
5th Author's Name Tetsunori KOBAYASHI
5th Author's Affiliation Department of Computer Science, Waseda University
Date 2003/11/14
Paper # PRMU2003-153,HIP2003-59
Volume (vol) vol.103
Number (no) 455
Page pp.pp.-
#Pages 6
Date of Issue