Music player control system using dynamic hand gesture recognition and position determination in 3D space

• Main Authors: Duka, Ivanhur O., Garcia, Lawrence G., Li, Jessica B., Sagabaen, Ruby Lynn
• Format: text
• Language: English
• Published: Animo Repository 2012
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/14782
• Institution: De La Salle University

The proliferation of digital music files nowadays results to users having problem with music selection. One of the possible keys is to classify these countless music files into genres. A recent study called Interactive 3D Music Organizer (i3DMO) made this possible however, the use of joystick and mouse to operate the software separately is another problem for the users. The Music Player Control System Using Dynamic Hand Gesture Recognition and Position Determination in 3D Music Organizer using hand gestures. It resolves the complications of using numerous buttons to control a music player system with its innovative user interface. In the system, both static and dynamic hand gestures are implemented. Algorithmic recognition is used to identify seven hand gestures, while Hidden Markov Models is used to recognize ten dynamic hand gestures. For this reason, Viterbi algorithm is used to obtain the best gesture model given a gesture pattern. These are grouped according to function (i3DMO-specific functions or music player controls), which can be used either by left hand or right hand. Each of these is based on 50 training samples and 100 test samples. Results show that the recognition rate per group is 93.35% and 94.864% for i3DMO-specific functions group and music player controls group respectively. User acceptability survey results show that the gestures are generally simple and easy to remember, and that the system is easy to use and responsive.