WEKO3
アイテム
Internal States Estimation Based on Facial Images by Separating Individual Features
http://hdl.handle.net/10228/0002001255
http://hdl.handle.net/10228/00020012557f82cef6-9596-43ea-b9a6-8d3c636f3354
| 名前 / ファイル | ライセンス | アクション |
|---|---|---|
10444571.pdf (5.2 MB)
|
|
| アイテムタイプ | 共通アイテムタイプ(1) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 公開日 | 2025-02-06 | |||||||||||
| タイトル | ||||||||||||
| タイトル | Internal States Estimation Based on Facial Images by Separating Individual Features | |||||||||||
| 言語 | en | |||||||||||
| 著者 |
Asaeda, Ayaka
× Asaeda, Ayaka
× 武村, 紀子
WEKO
35569
|
|||||||||||
| 抄録 | ||||||||||||
| 内容記述タイプ | Abstract | |||||||||||
| 内容記述 | Facial expressions are said to reflect a person’s internal state well and are commonly used in research on state estimation. However, individual facial features and expression differences can reduce state estimation accuracy. In particular, the estimation of ambiguous internal states such as concentration and drowsiness, which are complex to be expressed by facial expressions, are strongly affected by individual differences. Therefore, we aim to improve the estimation accuracy of ambiguous internal states by considering individual differences in facial images based on the Deviation Learning Network (DLN). In evaluation experiments, we estimate three levels of learners’ arousal (Awake/Drowsy/Asleep) using facial image data during e-learning. Furthermore, we used mixup to deal with the model’s ambiguity of class discrimination. | |||||||||||
| 言語 | en | |||||||||||
| 備考 | ||||||||||||
| 内容記述タイプ | Abstract | |||||||||||
| 内容記述 | 29th International Symposium on Artificial Life and Robotics, AROB 29th 2024, January 24-26, 2024, B-Con PLAZA, Beppu, JAPAN and ONLINE (Zoom) | |||||||||||
| 言語 | en | |||||||||||
| 書誌情報 |
en : Internal States Estimation Based on Facial Images by Separating Individual Features 発行日 2024-01 |
|||||||||||
| 言語 | ||||||||||||
| 言語 | eng | |||||||||||
| 資源タイプ | ||||||||||||
| 資源タイプ識別子 | http://purl.org/coar/resource_type/c_5794 | |||||||||||
| 資源タイプ | conference paper | |||||||||||
| 出版タイプ | ||||||||||||
| 出版タイプ | VoR | |||||||||||
| 出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||||||
| 会議記述 | ||||||||||||
| 会議名 | 29th International Symposium on Artificial Life and Robotics, AROB 29th 2024 | |||||||||||
| 言語 | en | |||||||||||
| 回次 | 29 | |||||||||||
| 開始年 | 2024 | |||||||||||
| 開始月 | 01 | |||||||||||
| 開始日 | 24 | |||||||||||
| 終了年 | 2024 | |||||||||||
| 終了月 | 01 | |||||||||||
| 終了日 | 26 | |||||||||||
| 開催国 | JPN | |||||||||||
| 研究者情報 | ||||||||||||
| URL | https://hyokadb02.jimu.kyutech.ac.jp/html/100001540_ja.html | |||||||||||
| 論文ID（連携） | ||||||||||||
| 値 | 10444571 | |||||||||||
| 連携ID | ||||||||||||
| 値 | 12690 | |||||||||||
