Research On Human-computer Interaction Model And Algorithm For Gesture Natural Interaction

With the continuous development of computer technology,the upgrading of wearing equipment and improvement of human's product experience requirements.Natural and harmonious interaction interface has become a hot topic in the field of interaction.The traditional "machine-centric" way of interaction has gradually shifted to "people-centered".In order to achieve people-centered intelligent interaction,the computer needs to perceive the user's operational intent,and then according to the perceived results,take the initiative to assist the user to complete the corresponding operation,thereby reducing the user's operating burden and understanding of the load.It is the key to realize the "people-centered" and the key research contents of this paper.This paper is supported by National Natural Science Foundation of China(No.61472163,61603151)National key R & D projects(2016YFB1001403)and Shandong Province key research and development projects(2015GGX101025).We use the visual gesture interaction for the virtual environment as the research background.Reducing the users' cognitive burdens and operating burdens,enhance the user experience as the goal.Perceive the user's intent of Virtual environment interaction interface,and interact according perception study results.In this paper,the main research content and innovation points are reflected as following aspects:(1)A dynamic gesture recognition method based on gesture key point change trend is proposedBy clockwise the gesture of the gesture,the number of fingers of the gesture and the position of the fingertip and the root of the gesture are determined.And the dynamic gesture is identified by the movement of the fingertip and the root.Contrast experiments show that the method has good robustness,and it is still better to recognize the effect when the segmentation is not very good,and the recognition result is not affected by the movement speed.For the usual nine dynamic gestures,the average recognition rate of this method is 96%.(2)A perceptual algorithm of perceived user operational intent is proposedBy analyzing the characteristics of the user's hand movement trajectory when the user controls the virtual man to select the object,the method of fitting the trajectory curve of the gesture by using the least squares method is proposed.According to the trend of the curve,the system initiatives move the predicted three-dimensional virtual object to the virtual hand;in addition,by dynamically adjusting the fitting data to improve the fitting effect and improve the correctness of the prediction results.(3)The human-computer interaction algorithm is proposedIn this paper,two kinds of perceptual interaction algorithms are proposed: motion-aware interaction algorithm and rotation-aware interaction algorithm.By recognizing the moving trajectory of the object,the system senses whether the user assembles the object.And the system updates the scene status according to perceived results,and realizes the motion-aware intelligent interaction;The system rotates the three-dimensional objects according to the clockwise or counterclockwise rotation gestures,and the result of the rotation is used as an implicit input to perceive the user's operational intent;finally the system according to the results of perception to make the appropriate operation to achieve the effect of rotating interaction.