Semantic Gesture Recognition Based on Cognitive Behavioral Model

Semantic gesture recognition based on video has an important research significance in the field of human-computer interaction. This paper proposes a semantic gesture recognition method based on the cognitive behavioral model. first, through analyzing the movement behavior of hand, we set up the transition probability matrix of semantic gestures. Second, set DDF-HMM model of semantic gesture. Then cognitive behavioral model is set up. According to the transition probability matrix, we could get the most likely semantic gesture of an unknown gesture. Then use the DDF-HMM model to identify semantic gesture. DDF-HMM method reduced gesture features´ dimension to a low level, then greatly reduced the complexity of the model. We recognized 4 kinds of semantic gestures, and the average recognition rate is 96.5%. The experimental results showed that this method can effectively distinguish different semantic gestures, and is more effective than Hu variant moments method.