A Decomposition Approach for Complex Gesture Recognition Using DTW and Prefix Tree
KAUST DepartmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Electrical Engineering Program
Permanent link to this recordhttp://hdl.handle.net/10754/656575
MetadataShow full item record
AbstractGestures are effective tools for expressing emotions and conveying information to the environment. Sequence matching and machine-learning based algorithm are two main methods to recognize continuous gestures. Machine-learning based recognition systems are not flexible to new gestures because the models have to be trained again. On the other hand, the computational time that matching methods required increases with the complexity and the class of the gestures. In this work, we propose a decomposition approach for complex gesture recognition utilizing DTW and prefix tree. This system can recognize 100 gestures with an accuracy of 97.38%.
CitationChen, H., Ballal, T., & Al-Naffouri, T. (2019). A Decomposition Approach for Complex Gesture Recognition Using DTW and Prefix Tree. 2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR). doi:10.1109/vr.2019.8797868
Conference/Event name2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)