Presentation | 2021-06-29 Investigation of pretext task for a classification model of human interaction motion Kenshiro Ata, Yusuke Nishimura, Yuya Okadome, Yutaka Nakamura, Hiroshi Ishiguro, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The technology of recognizing human actions has been used not only for recognizing human activities but also for controlling interactive robots in various situations. In this research, we focus on the task of recognizing human actions during an interaction. Although it is easy to record human actions during interaction using video equipment, it is costly to annotate the video data for training a discriminator. In this research, we develop an efficient learning method using self-supervised learning, in which unlabeled data is also used for learning the discriminant task. In this paper, we prepared several pretext tasks focusing on the temporal difference between the action of the dialogue partner and the task itself, and evaluated the difference in discrimination performance. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Interaction / Self-supervised learning / Pretext task / Motion recognition / Turn-taking |
Paper # | NC2021-14,IBISML2021-14 |
Date of Issue | 2021-06-21 (NC, IBISML) |
Conference Information | |
Committee | NC / IBISML / IPSJ-BIO / IPSJ-MPS |
---|---|
Conference Date | 2021/6/28(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Rieko Osu(Waseda Univ.) / Ichiro Takeuchi(Nagoya Inst. of Tech.) / 倉田 博之(九工大) / 関嶋 政和(東工大) |
Vice Chair | Hiroshi Yamakawa(Univ of Tokyo) / Masashi Sugiyama(Univ. of Tokyo) |
Secretary | Hiroshi Yamakawa(ATR) / Masashi Sugiyama(NICT) / (Univ. of Tokyo) / (AIST) |
Assistant | Nobuhiko Wagatsuma(Toho Univ.) / Tomoki Kurikawa(KMU) / Tomoharu Iwata(NTT) / Atsuyoshi Nakamura(Hokkaido Univ.) |
Paper Information | |
Registration To | Technical Committee on Neurocomputing / Technical Committee on Infomation-Based Induction Sciences and Machine Learning / Special Interest Group on Bioinformatics and Genomics / Special Interest Group on Mathematical Modeling and Problem Solving |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Investigation of pretext task for a classification model of human interaction motion |
Sub Title (in English) | |
Keyword(1) | Interaction |
Keyword(2) | Self-supervised learning |
Keyword(3) | Pretext task |
Keyword(4) | Motion recognition |
Keyword(5) | Turn-taking |
1st Author's Name | Kenshiro Ata |
1st Author's Affiliation | Osaka University(Osaka Univ.) |
2nd Author's Name | Yusuke Nishimura |
2nd Author's Affiliation | Osaka University(Osaka Univ.) |
3rd Author's Name | Yuya Okadome |
3rd Author's Affiliation | Institute of Physical and Chemical Research, Information R&D and Strategy Headquarters, Guardian Robot Project(RIKEN GRP) |
4th Author's Name | Yutaka Nakamura |
4th Author's Affiliation | Institute of Physical and Chemical Research, Information R&D and Strategy Headquarters, Guardian Robot Project(RIKEN GRP) |
5th Author's Name | Hiroshi Ishiguro |
5th Author's Affiliation | Osaka University(Osaka Univ.) |
Date | 2021-06-29 |
Paper # | NC2021-14,IBISML2021-14 |
Volume (vol) | vol.121 |
Number (no) | NC-79,IBISML-80 |
Page | pp.pp.97-102(NC), pp.97-102(IBISML), |
#Pages | 6 |
Date of Issue | 2021-06-21 (NC, IBISML) |