Faculty of Computer Science - Universitas Indonesia
Language
eng
Abstract
The main target in this research is a design of a virtual martial arts training system in real-time and as a tool in learning martial arts independently using genetic algorithm methods and dynamic time warping. In this paper, it is still in the initial stages, which is focused on taking data sets of martial arts warriors using 3D animation and the Kinect sensor cameras, there are 2 warriors x 8 moves x 596 cases/gesture = 9,536 cases. Gesture Recognition Studies are usually distinguished: body gesture and hand and arm gesture, head and face gesture, and, all three can be studied simultaneously in martial arts pencak silat, using martial arts stance detection with scoring methods. Silat movement data is recorded in the form of oni files using the OpenNI ™ (OFW) framework and BVH (Bio Vision Hierarchical) files as well as plug-in support software on Mocap devices. Responsiveness is a measure of time responding to interruptions, and is critical because the system must be able to meet the demand.
Collaboration network of the authors
7 authors ·
21 co-authorship links
· 11 across different organisations
Hover or tap a name to highlight its co-authors. Double-click to open the author's page.
Universitas Sintuwu Maroso
Universitas Jayabaya
Gunadarma University
Affiliation unknown
Different organisations
Same organisation
Affiliation unknown
Thicker line = more joint articles · bigger circle = more articles
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.