WEKO3
アイテム
ストリートダンスの振り付け支援のためのLLMを用いた位置構成生成手法の試作
https://ipsj.ixsq.nii.ac.jp/records/231625
https://ipsj.ixsq.nii.ac.jp/records/231625acd4b52e-84df-415a-87d1-b483024863ce
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
IPSJ-HCI24206024.pdf (4.0 MB)
2026年1月8日からダウンロード可能です。
|
Copyright (c) 2024 by the Information Processing Society of Japan
|
|
| 非会員：¥660, IPSJ:学会員：¥330, HCI:会員：¥0, DLIB:会員：¥0 | ||
| Item type | SIG Technical Reports(1) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2024-01-08 | |||||||||
| タイトル | ||||||||||
| タイトル | ストリートダンスの振り付け支援のためのLLMを用いた位置構成生成手法の試作 | |||||||||
| タイトル | ||||||||||
| 言語 | en | |||||||||
| タイトル | Prototype of LLM-based Positional Configuration Generation Method for Street Dance Choreography Support | |||||||||
| 言語 | ||||||||||
| 言語 | jpn | |||||||||
| キーワード | ||||||||||
| 主題Scheme | Other | |||||||||
| 主題 | スポーツ・動作 | |||||||||
| 資源タイプ | ||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_18gh | |||||||||
| 資源タイプ | technical report | |||||||||
| 著者所属 | ||||||||||
| 名古屋工業大学大学院工学研究科情報工学専攻 | ||||||||||
| 著者所属 | ||||||||||
| 名古屋工業大学大学院工学研究科情報工学専攻 | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Department of Computer Science and Engineering, Graduate School of Engineering, Nagoya Institute of Technology | ||||||||||
| 著者所属(英) | ||||||||||
| en | ||||||||||
| Department of Computer Science and Engineering, Graduate School of Engineering, Nagoya Institute of Technology | ||||||||||
| 著者名 |
杉山, 紘次郎
× 杉山, 紘次郎
× 白松, 俊
|
|||||||||
| 論文抄録 | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | When a team dances on the street, they often decide by trial and error how they will move and what kind of positional configurations they will make while dancing. In this study, we developed a prototype method for generating positional configurations using a large-scale language model (LLM) in order to support such a co-creative consensus building process. Specifically, a model is constructed using GPT-3.5 fine-tuning, which takes 8 bars of music information, concepts to be expressed, genres, number of people, and other assumptions as inputs, and outputs positional configurations and dance descriptions. For this purpose, we constructed training data by dividing showcases actually danced at JAPAN DANCE DILIGHT[1] into music information, positional configuration coordinates, and dance outlines. The system proposes three patterns generated every eight measures, and the user selects the recommended positional configuration while discussing it with other dancers. By repeating this process, the system aims to provide support for dance choreography. As an evaluation experiment, we will conduct a comparison of user satisfaction and usability with the case in which the system is not used. | |||||||||
| 論文抄録(英) | ||||||||||
| 内容記述タイプ | Other | |||||||||
| 内容記述 | When a team dances on the street, they often decide by trial and error how they will move and what kind of positional configurations they will make while dancing. In this study, we developed a prototype method for generating positional configurations using a large-scale language model (LLM) in order to support such a co-creative consensus building process. Specifically, a model is constructed using GPT-3.5 fine-tuning, which takes 8 bars of music information, concepts to be expressed, genres, number of people, and other assumptions as inputs, and outputs positional configurations and dance descriptions. For this purpose, we constructed training data by dividing showcases actually danced at JAPAN DANCE DILIGHT[1] into music information, positional configuration coordinates, and dance outlines. The system proposes three patterns generated every eight measures, and the user selects the recommended positional configuration while discussing it with other dancers. By repeating this process, the system aims to provide support for dance choreography. As an evaluation experiment, we will conduct a comparison of user satisfaction and usability with the case in which the system is not used. | |||||||||
| 書誌レコードID | ||||||||||
| 収録物識別子タイプ | NCID | |||||||||
| 収録物識別子 | AA1221543X | |||||||||
| 書誌情報 |
研究報告ヒューマンコンピュータインタラクション（HCI） 巻 2024-HCI-206, 号 24, p. 1-6, 発行日 2024-01-08 |
|||||||||
| ISSN | ||||||||||
| 収録物識別子タイプ | ISSN | |||||||||
| 収録物識別子 | 2188-8760 | |||||||||
| Notice | ||||||||||
| SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. | ||||||||||
| 出版者 | ||||||||||
| 言語 | ja | |||||||||
| 出版者 | 情報処理学会 | |||||||||
