學術資源整合系統-相關推薦
 |
|
|
|
|
| 作者 | Huang, Chun-Yen;Chou, Yin-Yu;Yan, Yi-Kuei;Chiu, Yu-Hua |
|---|
| 出版日期 | 20260708 |
|---|
| 著作名稱 | A Human Centric IoT Exhibition Learning System for Museums Integrating Gesture Recognition and LLM Based Information Services
|
|---|
| 會議名稱 | 2026 8th Blockchain and Internet of Things Conference |
|---|
| 會議地點 | Tokyo, Japan |
|---|
| 主辦單位 | BIOTC 2026 |
|---|
| 國際性會議 | Y |
|---|
| 其他資訊 | EI |
|---|
| 主題 | 資訊科學;展示 |
|---|
| 關鍵字 | Human-Centric IoT, Gesture Recognition, Large Language Models, Transpar-ent Display, Interactive Exhibition, Museum Information Services. |
|---|
| 摘要 | This study proposes an IoT-enabled transparent interactive showcase for pale-ontological learning in a natural science museum. The system integrates visual sensing, edge-based gesture recognition, and cloud-based large language model (LLM) services to support real-time, context-aware interaction. When no visi-tor is detected, the transparent display remains clear, allowing unobstructed observation of the fossil specimen. Once a visitor is detected, the system over-lays basic information, such as the fossil name, on the display. Visitors can then perform predefined numeric hand gestures, which are recognized at the edge device and translated into interaction commands. These commands are transmitted to a cloud-based LLM service to generate detailed fossil infor-mation in real time, and the results are rendered on the transparent interface. From an IoT perspective, the system represents a sensing-to-service architec-ture that combines sensor input, edge computing, and cloud-based AI services to deliver adaptive information in a smart public environment. A preliminary evaluation involving 31 participants indicated positive perceived usability and satisfaction, with a mean score of 4.17 for the directionally recoded SUS items and 4.12 for overall satisfaction on a five-point scale. The proposed approach demonstrates how museum exhibits can be transformed into human-centric IoT applications that provide seamless integration of physical observation and dy-namic digital learning. |
|---|
| 系統號 | NO000007737 |
|---|
|
|