透過您的圖書館登入
IP:18.117.76.7
  • 學位論文

使用視訊穿透式頭戴顯示裝置應用於太極拳學習之全身動作引導

Full-Body Movement Guidance For Learning Tai Chi Chuan With A Video See-Through Head-Mounted Display.

指導教授 : 洪一平

摘要


在現今高齡化社會中,良好的運動習慣備受重視,而科技的進步使得人們能在家中運用多媒體設備來學習各種運動,像是透過教學影片來學習太極拳動作。藉由此方式,使用者雖然可觀看太極拳老師的動作精華,但多媒體影片只有平面的影像資訊,對於空間位置資訊是缺乏的。除此之外,本身並沒有任何回饋資訊,導致使用者無法知道自己的動作是否做正確。在傳統的運動訓練班中,動作的學習會經由教練的指導與回饋來提升其正確性。在沒有教練的情況下,學員很難自行學習正確動作。因此本研究透過結合視訊穿透式頭戴顯示器與深度攝影機來提供使用者以第一人稱的方式來學習太極拳,其中包含骨架疊合之校正方法,並在擴增實境(Augmented Reality)中,設計兩種不同的引導方式。其一為教練環顧引導,在使用者的四周圍擺放虛擬教練來讓使用者可以觀看不同角度的虛擬教練所打出的動作來學習,藉此獲得更多的角度資訊來達到動作的理解性;其二為跟隨球引導,我們在使用者預做動作的關節點上擺放一顆虛擬球,並讓使用者跟隨AR環境中的虛擬球,來提升動作之正確性。 最後本研究透過姿勢、動作與太極拳案例三種不同使用者研究,並以抽象動作、運動動作及太極拳動作來進行實驗。在姿勢(Posture)與動作(Movement)的研究結果共同顯示出,混合教練環顧與跟隨球引導可提升姿勢與動作之正確性外,在身體動作的理解上也較容易了解;其中使用者在姿勢的表現上整體平均可在 0.142公尺差異內完成;動作的表現上整體平均可在 0.144公尺差異內完成。

並列摘要


People are paying more and more attention to having good exercise habits in this aging society. The advancement of technology grants the possibility of learning various kinds of exercises using various multi-media equipment. For instance, Tai Chi Chuan movements can be learned through watching instruction videos. Through this method, users can examine the teacher demonstrating most important parts of movements. However, multi-media videos not only lack feedback information but also only let users understand the movements by showing the plane images. In a traditional exercise class, the movements of students are taught by a teacher directly. Without a teacher, students would have difficulty learning movements by themselves. This research provides users a way in the first-person point of view to learn Tai Chi Chuan through combining the video see-through head-mounted display and depth camera. It contains the method for skeleton calibration and two different kinds of guidance are designed to carry out in Augmented Reality (AR). One is Coach-Surrounding guidance. In this approach, virtual coaches are set surrounding the user. Virtual coaches allow the user to see demonstration movement from different viewing angles. Thus users can acquire more visual information of the movement which allow them to gain a clearer understanding. The other is Ball-Following guidance. In this approach, we set virtual balls on certain joint points and require user to follow virtual balls in Augmented Reality (AR) in order to promote the accuracy of movements. Finally, we conducted three user studies-a posture study, a movement study, and a Tai Chi Chuan study. Both posture and movement studies show common results that combined Coach-Surrounding and Ball-Following guidance can promote the accuracy of movements. In addition, it is easier for users to understand the physical action. Exceeding our expectations, participants performed postures with an average error of 0.142m, and performed movement with an average error of 0.144m.

參考文獻


[1] CMU Graphics Lab Motion Capture Database: http://mocap.cs.cmu.edu.
[3] O. Portillo-Rodriguez, O. O. Sandoval-Gonzalez, E. Ruffaldi, R. Leonardi, C. A. Avizzano, and M. Bergamasco. Real-Time Gesture Recognition, Evaluation and Feed-Forward Correction of a Multimodal Tai-Chi Platform. Haptic and Audio Interaction Design, 5270:30–39, 2008.
[4] Anderson, F., Grossman, T., Matejka, J., & Fitzmaurice, G. YouMove: enhancing movement training with an augmented reality mirror. In ACM UIST, 311-320, 2013.
[5] Jacky CP Chan, Howard Leung, Jeff KT Tang, and Taku Komura. A virtual reality dance training system using motion capture technology. Learning Technologies, IEEE Transactions on, 4(2):187–195, 2011.
[6] Kozaburo Hachimura, Hiromu Kato, and Hideyuki Tamura. A prototype dance training support system with motion capture and mixed reality technologies. In Robot and Human Interactive Communication, 2004. ROMAN 2004. 13th IEEE International Workshop on, pages 217–222. IEEE, 2004.

延伸閱讀