Personalized Gesture Recognition and Its Applications

In “uWave: Accelerometer-based Personalized Gesture Recognition and Its Applications ,” Jiayang Liu, Zhen Wang, and Lin Zhong at Rice University in Houston, Texas, and Jehan Wickramasuriya and Venu Vasudevan at the Motorola Labs Applications & Software Research Center have developed an efficient recognition algorithm for such gesture recognition and interaction using a single three-axis accelerometer.

Unlike statistical methods, their uWave system only requires a single training sample for each gesture pattern and allows users to employ personalized gestures and physical manipulations. They evaluated uWave using a large gesture library with over 4000 samples collected from eight users.

Based on a “vocabulary” of eight gesture patterns, they used the algorithm to to achieve 98.6% accuracy, competitive with statistical methods that require significantly more training samples.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.