姿勢是一種包含信息的身體動作，姿勢辨識技術是透過影像辨識技術來偵測人的各種姿勢變化進而提供互動回饋，本研究期望應用此技術在互動學習領域，透過於空間環境中「做中學」的概念，讓使用者以身體姿勢無負擔的方式與學習內容產生互動，讓使用者能夠融入系統所建構的環境或情境並得到適當的學習回饋。
在本研究中，作者將以錯視覺及敦煌連環壁畫作為互動學習之內容。分別以偵測手揮動以及腳步移動姿勢來設計兩套裝置式互動學習系統：「錯視覺手揮動互動學習牆」與「敦煌連環壁畫互動學習走道」。結合空間裝置藝術概念與互動媒體技術，兩套系統的設計理念均是讓參與者參與學習，盡量符合自然互動的精神。希望本論文之應用經驗未來能提供更多教學單位、數位藝術創作者或博物館參考使用。
關鍵詞：姿勢辨識、自然互動、數位學習、裝置式互動學習、未來博物館、數位藝術、錯視覺、敦煌連環壁畫
Gesture is one kind of message to body movement. The gesture-based technology detects various human beings’ gesture through image recognition technology and provides interactive feedback. The purpose of this study is to apply this technology in interactive learning system and in the concept of learning in “space installation” to make participants interact naturally in the systems and to allow users to immerse themselves in the learning system and to get the appropriate feedback from learning.
In this thesis, we used “Visual illusion artistic” and “Dunhuang running fresco” as the context of learning topics in our system. By using natural movements such as “walking” or “hand gestures”, we designed two interactive learning systems: “Walking through Dunhuang running fresco” and “Using hand movements to control optical illusions”. Integrating with the form of installation art and interactive media technology, the basic idea of the two systems is to conform to natural interaction essence. We hope that in the future the application in our systems can be used as a reference of digital artists and learning experts.
Keywords：Gesture Recognition, Nature Interaction, Digital Learning, Interactive Learning with Installation, Future Museum, Digital Art, Visual Illusion, Dunhuang Running Fresco