3D keyframe motion extraction from zapin traditional dance videos

• Main Authors: Albakri, Ikmal Faiq, Nik Wafiy, Nik Wafiy, Mohd. Suaib, Norhaida, Mohd. Rahim, Mohd. Shafry, Yu, Hongchuan
• Format: Conference or Workshop Item
• Published: 2020
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://eprints.utm.my/id/eprint/89769/; http://dx.doi.org/10.1007/978-981-15-0058-9_7
• Institution: Universiti Teknologi Malaysia

Capturing traditional dance motion data using motion capture technology is too expensive as it required a lot of high-priced equipment involving complicated procedures. It is also hard to find any traditional dance motion data from any open sources as the motion data always being made private and not for public use. However, the traditional dance videos can be found easily from any internet medium such as YouTube and it also free to be used by the public. Therefore, in this paper we propose a method of extracting the 3D dance motion data of Zapin traditional dance from videos instead of motion capture technology by using keyframe animation extraction method. This method works by extracting all the 3D body joints coordinates of every frame in the keyframe dance animation and reconstruct the extracted motion data into an empty character model to check whether the extracted and reconstructed 3D Zapin dance motion data is correct and efficient compare to the original dance motion in the videos.