Presentation | 2022-03-03 A Study of Features for Estimating Attitude of Participation in Discussion from Audio and Video Tsubasa Kanaoka, Sunao Hara, Masanobu Abe, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The purpose of this report is to detect students who have poor engagement in class discussions. In this paper, we recorded a total of about 3 hours of discussions in 3 groups to be used for estimating the engagement. And, Using the features extracted from the audio and video, we generated a random forest to estimate the binary posture labels, and analyzed the importance. Through the analysis of the importance, we examined the features necessary for the estimation of the engagement. In the analysis of importance, for acoustic features, the spectrum in the low frequency band that represented the amount of speech, and the logarithmic energy of the sound that represented the excitement of the discussion were important. In the case of video features, vertical motion features that represented writing and nodding and action unit that represented facial expression change were important. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | OpenFace / OpenPose / Multiplayer Interaction / Discussion Analysis / Random Forest end{ekeyword} |
Paper # | LOIS2021-49 |
Date of Issue | 2022-02-24 (LOIS) |
Conference Information | |
Committee | LOIS |
---|---|
Conference Date | 2022/3/3(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Toru Kobayashi(Nagasaki Univ.) |
Vice Chair | Hiroyuki Toda(NTT) |
Secretary | Hiroyuki Toda(Nagasaki Univ.) |
Assistant | Kazuki Fukae(Nagasaki Univ.) |
Paper Information | |
Registration To | Technical Committee on Life Intelligence and Office Information Systems |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | A Study of Features for Estimating Attitude of Participation in Discussion from Audio and Video |
Sub Title (in English) | |
Keyword(1) | OpenFace |
Keyword(2) | OpenPose |
Keyword(3) | Multiplayer Interaction |
Keyword(4) | Discussion Analysis |
Keyword(5) | Random Forest end{ekeyword} |
1st Author's Name | Tsubasa Kanaoka |
1st Author's Affiliation | Okayama University(Okayama Univ.) |
2nd Author's Name | Sunao Hara |
2nd Author's Affiliation | Okayama University(Okayama Univ.) |
3rd Author's Name | Masanobu Abe |
3rd Author's Affiliation | Okayama University(Okayama Univ.) |
Date | 2022-03-03 |
Paper # | LOIS2021-49 |
Volume (vol) | vol.121 |
Number (no) | LOIS-401 |
Page | pp.pp.57-62(LOIS), |
#Pages | 6 |
Date of Issue | 2022-02-24 (LOIS) |