應用擴增實境擷取小肢體動作之研究

Small Motor Motion Capture for Augmented Reality Applications

張琴翊(Chin-Yi Chang)

指導教授 ： 鄭宗明

朝陽科技大學/理工學院/工業工程與管理系/碩士(2015年)

摘要

擴增實境(Augmented Reality, AR)是將實境背景影像中重疊呈現虛擬物體影像之技術，藉由實境中之標記(Marker)即時追蹤與方位辨識來呈現相對應之虛擬物體外觀，以傳遞虛實合一之概念。如此之方法，可以將不存在於背景實境的物體，模擬出靜態或動態之虛實互動效果，藉此實施研究或規劃。由於手部動作精緻且靈巧，完整的動作擷取並非易事且未必需要。然而，使用昂貴之穿戴設備擷取動作資訊會增加手部的負擔，讓工作感受失真。若是只追蹤指尖動向而將動作簡化處理，亦存在著動作資訊遺漏的問題。
本研究以手部於桌面上之工作為探討範圍，藉由貼附AR標記於關鍵之手部位置，再使用擴增實境方法追蹤手部動作。雖然於擴增實境空間中之座標值很可能與實體直角空間之座標值不同，但其間仍保有一對一的對應關係。因此，將AR座標值轉換後即可獲得其真實空間值，藉此將手部動作與虛擬空間之物體進行互動，並取得互動數據，進行各式虛實互動之分析研究。

關鍵字：擴增實境、虛擬實境、虛實互動、手部動作擷取

關鍵字

手部動作擷取 ； 虛實互動 ； 虛擬實境 ； 擴增實境

並列摘要

The techniques in Augmented Reality (AR) can superimpose realistic 3D image of the virtual subject onto a video-based background to create a reasonable fusion of the intended purpose. In the method, the setting of the markers plays an important role for a successful fusion in terms of speed response and positioning. However, the motion of a human hand is quick and subtle, and was usually captured by uncomfortable electronic wearing device. Some methods traced only the displacements of the fingertips and then simulate the motions of the whole palm. In order to find the balance between the uncomfortable heavy wearing and potential data lose, this research placed markers on a human hand and then studies the performance of its virtual counterpart.
The setup in this research considered only the motions performed on a desk top, and the AR markers were placed only on the essential portions of a hand for which would lower computational loading and were sufficient to represent an average working hand. The major computation would be on the coordinate calibration between the 3D space and the AR space. Several operations were tested to determine the performance of the methodology.

Keywords: Augmented Reality, Virtual Reality, Virtual-Physical Interaction, Hand Motion Capture

並列關鍵字

Virtual-Physical Interaction ； Virtual Reality ； Augmented Reality ； Hand Motion Capture

參考文獻

Athitsos, V., Sclaroff, S., An Appearance-Based Framework for 3D Hand Shape Classification and Camera Viewpoint Estimation, IEEE International Conference on Automatic Face and Gesture Recognition, pp. 45-50, 2002.

Azuma., A Survey of Augmented Reality, Teleoperators and Virual Environments, pp. 355-385, 1997.

Bimber, O., Raskar, R., Spatial Augmented Reality Merging Real and Virtual Worlds, 2005.

Dawson, C W. and Wilby, R. L.,Hydrological Modeling Using Artificial Neural Networks. Progress in Physical Geography, Vol.25(1), pp. 80-108, 2001

Fun, M., Hagan, M.T., Levenberg-Marquardt Training for Modular Networks, IEEE International Conference, Vol.1, pp. 468-473, 1996.