| 研究課題/領域番号 |
23K21688
|
|---|
| 補助金の研究課題番号 |
21H03482 (2021-2023)
|
|---|
| 研究種目 |
基盤研究(B)
|
| 配分区分 | 基金 (2024)
補助金 (2021-2023) |
|---|
| 応募区分 | 一般 |
|---|
| 審査区分 |
小区分61020:ヒューマンインタフェースおよびインタラクション関連
|
|---|
| 研究機関 | 大阪大学 |
|---|
研究代表者 |
ORLOSKY JASON 大阪大学, サイバーメディアセンター, 特任准教授(常勤) (10815111)
|
|---|
| 研究分担者 |
白井 詩沙香 大阪大学, サイバーメディアセンター, 講師 (30757430)
清川 清 奈良先端科学技術大学院大学, 先端科学技術研究科, 教授 (60358869)
|
|---|
| 研究期間 (年度) |
2021-04-01 – 2025-03-31
|
| 研究課題ステータス |
交付 (2024年度)
|
| 配分額 *注記 |
17,160千円 (直接経費: 13,200千円、間接経費: 3,960千円)
2024年度: 1,430千円 (直接経費: 1,100千円、間接経費: 330千円)
2023年度: 3,120千円 (直接経費: 2,400千円、間接経費: 720千円)
2022年度: 5,720千円 (直接経費: 4,400千円、間接経費: 1,320千円)
2021年度: 6,890千円 (直接経費: 5,300千円、間接経費: 1,590千円)
|
|---|
| キーワード | Eye Tracking / Artificial Intelligence / Learning / Agents / augmented reality / virtual reality / eye tracking / State Detection / Remote Environments / agents / cognition / learning / simulation / remote interaction / education |
|---|
| 研究開始時の研究の概要 |
This research involves integrating voice-to-text with AR for interactive learning, customizing LLMs for conversational education styles, and refining these technologies through user feedback. Evaluations will be conducted for enhanced language and content learning.
|
| 研究実績の概要 |
In this period, we explored applications of virtual reality (VR) and augmented reality (AR) in education and healthcare. First, we conducted a study on educational comics that utilized eye-tracking in VR to identify key gaze features to help estimate the difficulty levels perceived by readers, suggesting a way to dynamically adjust educational content. We also developed, AMSwipe, a method that allows for gaze-based text input into virtual environments, which allows for efficient, hands-free typing without the need for physical controllers. Additionally, EyeShadows, a tool that we developed and tested with both AR and VR, improves the selection and manipulation of virtual elements using peripheral copies of items, enabling faster, more accurate interactions. Furthermore, we leveraged VR to enhance medical training, particularly by simulating the experiences of Parkinson’s disease patients to foster empathy and insight among healthcare students. These technologies demonstrate significant potential for enhancing remote education, providing immersive, interactive learning experiences that can be tailored to individual needs and capabilities. We also explored the use of interactive re-training of neural networks for applications in language learning.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
This research is generally progressing on schedule. We have several publications in different areas, and we have set up a remote environment to conduct the last phase of the research over the next year.
|
| 今後の研究の推進方策 |
The last phase of the research will proceed as planned, though we have made some updates due to the advancement of AI technology. In addition, we are working on an in-situ object labeling approach, which can assist with a more specific learning task, language learning. Our system will be extended to incorporate Large Language Models(LLM), which will be customized to power virtual educational agents. This should result in more interactive and context-based learning, leading to longer retention of the concepts.
|
