Abstract
Traditional dance in Malaysia is generating considerable amount of interest due to its unique elements of heritage which have contributed to its diverse music and dance forms. For example, Zapin, Kuda Kepang, Mak Yong, Joget, Ngajat and much more. Recent developments in technology and ever- growing online community, traditional dance are undergoing a revolution where these dance form can be studied and observed easily especially when there are dance software that can help guide users to learn by performing the dance steps in real-time. However, the use of gesture sensor for accurately mapping the dance movements of traditional dance is not yet explored, since only modern dances are normally available to the masses in the form of computer games. This paper outlines a new approach to implement Normalize Dynamic Time Warping (NDTW) algorithm using skeleton tracking techniques to imitate the intricate movements of traditional dance and to assess the robustness of the algorithm. For this study, the traditional dance of Zapin was chosen because it consists of simple body movements and data were acquired using Microsoft Kinect. The results showed that the proposed algorithm gave the overall matching rate of 99.21% with maximum mean success rate of dancers gave 99.68% and non-dancers gave the percentage of 98.76%. This technique may be considered as a relatively unexplored application area, and the proposed system is an attempt to address the problem with reasonable accuracy and scopes for further research.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call