結合體感辨識與觸發器運用於角色動畫之研究

A Study of Combining Somatosensory Technology and Trigger Applied in Character Animation

近年來3D體感辨識技術（Somatosensory Technology）快速的發展，使其能偵測、定位使用者的一舉一動。然而此技術應用極為廣泛，在電腦遊戲方面創造出人機互動（Human-Computer Interaction）的新體驗；在教學訓練方面有肢體動作的學習；在醫療復健方面有中風患者及外科手術的遠端指令遙控運用等。本文擬藉由體感辨識技術以及觸發器（Trigger）的設置，完成播放虛擬化身（Avatar）之混合動畫。在系統實作上是應用體感裝置感應使用者的動作姿態，將其轉換成關節圖（Joint-Figure）與深度圖（Depth-Figure）套用於系統中的人體模型，再比照人體的各關節部位配置感應點於人體模型上，並結合混合動畫之設計，使得虛擬化身能依照人體模型碰觸到的觸發器，完成播放相對應的混合動畫。此體感辨識混合動畫之執行，系統是運用間接操控的方式，播放虛擬化身的角色動畫，其優點是使用者不需要持續動作來操控角色動畫，只需要碰觸到觸發器即可播放角色動畫，本系統亦可運用於3D遊戲系統中，使得體感辨識技術帶來更多樣化的互動與便利性。

3D Somatosensory Technology which detects and locates every single movement of the user’s, is booming in recent years. The technique has been widely applied, particularly in games where you can have a whole new experience of human-computer interaction. Furthermore, it can also be applied in teaching where limbs movement is involved and in medical treatment such as rehabilitation for stroke patients and remote control instructions utilized in surgical operation. The somatosensory technology is set up by using a trigger to play the avatar in blending character animation in the study. Practically, a somatosensory device is employed to detect user’s gesture which is converted to the joint-figure and depth-figure in the human body modeling. Then, the avatar can play the corresponding blending animation according to the triggers on the human body modeling which reflects human’s joints being activated incorporating with the design of the blending animation. The switch among somatosensory blending character animation is executed by an indirect control to play character animation of avatar. The advantage of it is that users do not need to move continuously to manipulate character animation but only need to touch the trigger to play. The system can also be utilized in 3D games where a somatosensory technology can bring more diverse interactions and applications.