Research On Correlation Sensing Method Of The Scene Interaction Event For Smart TV

With the development of the theory and technology of computer, Internet and Pervasive computing, people's life has been deeply influenced. Meanwhile, human-computer interaction technology has become more and more integrated into people's work and life. With the development of society and the progress of science and technology, the traditional human-computer interaction technology already cannot satisfy people's needs. However, natural human-computer interaction has been paid more and more attention by people, and it is natural, harmonious, intelligent and efficient. As a kind of natural interactive medium, gesture has the abundant, natural and intuitive characteristics of expression. Therefore, gesture interaction has become an important research topic in the field of human-centered natural human computer interaction.The natural gesture interaction based on vision is natural, convenient and is consistent with the characteristics of people's daily communication habits. It has been applied in sign language interaction, robot control, smart-Home interaction, interactive games and other fields, and it has wide application prospect. But in the gesture interaction based on vision, there are still problems of the low interactive intelligence, the Midas Touch Problem, user's heavy cognitive load and heavy load. The realization of the interaction is purely dependent on the pattern recognition rate of gesture. This severely limits the application and User Experience of the gesture interaction based on vision in the practical work and life. This paper has been funded from the National Natural Science Foundation of China(No.61173079, No.61472163) and Key Project of Natural Science Foundation of Shandong Province(No.ZR2011FZ003). We use the visual gesture interaction for smart TV as the research background, and we researched the key problems in the visual gesture interaction from the perspective of user's intention perception based on interaction context and implicit human-computer interaction. The main research contents and innovations of this paper are as follows:(1) According to the characteristics of interaction and interactive scene of smart TV, we proposed the method of gesture segmentation and the dynamic gesture recognition algorithm. And it meets the requirements of the vision gesture interaction for smart TV.(2) In this paper, we analyzed the interactive scene of visual gesture interaction for smart TV from the perspective of cognitive psychology and cognitive behavior. On this basis, we established a multi-level dynamic interactive context model which was based on user's behavior and smart TV's functional state, and a visual single gesture interaction prototype system for smart TV which was based on the interactive context. They effectively reduce the operation load and the cognitive load of user.(3) We proposed the inference logic model of graph and the inference model of the dynamic context(CDL-DFCM). And they can identify the situational event of interaction and perceive the user's intention.(4) We proposed an implicit interaction model which was based on the data fusion of the explicit information and implicit information, the explicit and implicit interaction algorithm. The method improves the operation accuracy and user experience in the interaction between the user and smart TV.