Online Classroom Virtual Facial Expression-driven Application Research

Virtual facial expression drive is to generate facial expression animation by using3 D face model.This article mainly studies the designing of virtual facial expression drive application system of online classroom based on virtual facial expression drive captured by facial expression,which applies the virtual facial express drive to the online classroom and improves the classroom atmosphere and teaching effect.Focus on the system design facial expression capture and facial expression animation generation,this main contents of this article are as follows:1.Design video face feature point tracking algorithm.Collect the facial expression image with the monocular camera.Then extract face detection and feature points with multi task learning cascade convolution neural network algorithm and the cascade gradient enhanced regression tree algorithm,and add a time threshold processing mechanism for correlation filtering algorithm to track the face detection frame,realize the face feature point tracking in video,and prove the stability and real-time of this algorithm by experiment.2.Facial expression animation control parameter includes head posture and facial expression parameters.The head pose is solved by EPn P algorithm,and output head rotation matrix and translation vector.Propose to use and support vector machine span date sets learn from AU detector and strength value regressor,in order to deal with the problem of the lack of labeling of facial movement unit AU data set.And obtain the facial expression parameters by establishing the mapping relationship between AU and blendshape expression.3.Design the virtual facial expression driving application system of online classroom,drive Blendshape animation model to generate facial expression animation by transmitting Unity3 D software from facial expression control parameter,and display the effect of facial expression animation in online classroom through virtual camera driving software at last.