The Gesture Semantic Classification Algorithm for Reducing the User's Memory Load

Abstract

Aiming at reducing the user memory burden , a method is proposed that started with the multi-semantics of the same gesture , and was based on Gaussian statistical model for gesture semantic classification.After implementing the cognitive observation and experiment , a flexible mapping algorithm from gesture to semantics and using Gaussian statistical model was proposed . The main innovations of this paper are as follows: (1) A mapping mechanism for multi-semantics with the same gesture to reduce the users memory load and its interaction algorithm are proposed. (2) Using the Gaussian statistical model analyzes the features of different semantics under the same gesture, and the parameter learning of the Gaussian model called the EM method makes it possible to establish a more stable correspondence between the gesture semantics and the motion information, and to identify the users intention more accurately. At the same time,the application of the algorithm on operating the geometry in the intelligent classroom was implemented. The results showed that the algorithm could accurately classify the gesture semantics, express the intention of the user and reduce the users memory burden.