Intelligent Multimedia Teaching Demonstration System Based On Kinect

With the development of science and technology,multimedia technology has become an indispensable part of people's daily life and work.The traditional multimedia technology is complex,poor flexibility,poor portability,and cannot meet the requirements of people.Thus seeking more intelligent multimedia control system has become a key.The Kinect sensor which is more flexible and natural has changed the traditional way of human-computer interaction,so the Kinect sensor combined with multimedia technology has tremendous application value.In this paper,the face recognition and gesture recognition of multi-media teaching demonstration system are researched;the main contents are as follows:First,this paper introduces the technology of human-computer interaction and the present situation of the teaching demonstration system,then give a brief of the research status of face recognition and gesture recognition at home and abroad.The hardware structure and related parameters of Kinect are summarized,and the data acquisition of color image,depth image and bone tracking is realized.Second,a face recognition system based on Kinect is constructed.Several image preprocessing methods are researched and realized,the face detection and orientation of the real-time video images is realized.The principal component analysis method is used for the face recognition,and also this paper analysis the recognition effect under different light conditions of self-face database.Then the application of face recognition system in teaching demonstration system is used for opening PPT files.Third,use the Kinect sensor to track the gesture of human body.The characteristics of the gesture are analyzed and the gesture trajectory is extracted.Then this paper defines four kinds of gestures which are used in the system,and analysis of gesture recognition in different light conditions and distance recognition results.At last,the hand gesture recognition is used to control the PPT document.Finally,the system software design is realized.The establishment of self-face database and model training file creation are achieved.Afterwards,the gesture recognition front-end interface is designed,then gives the flow chart of the system and implements the system as a whole.At last,this paper verifies the feasibility of the system.