Research On Real-time Facial Animation Interactive Technology Based On Monocular Camera

As an important carrier of human information and emotional expression,the human face is an important component of the new generation of human-computer interaction.Face animation has been widely used in industry,but traditional face animation production methods are time-consuming,labor-intensive,and costly.Therefore,it is of great significance to find a simple,convenient and low-cost face animation production method.Aiming at the problems of face animation technology,this paper conducts in-depth research and discussion around the face real-time interactive animation technology based on monocular video cameras.Our research work covers three core parts of facial animation technology:1.Facial expression capture and tracking.The facial expression tracking and capture mainly include two parts: face detection algorithm and facial feature point location algorithm.This paper proposes a face detection method using a directional gradient histogram(HOG)and a support vector machine(SVM).First,a monocular video camera is used as a collection device for face video,and then the HOG features of the face image are extracted.The SVM learns the HOG features to select the support vector to construct the optimal classification surface.The face area in the video frame is detected according to the trained SVM classifier.After obtaining the face area,we use the cascade regression tree-based face feature point location algorithm to locate the feature points of key parts of the face(eyebrows,mouth corners,nose tip,etc.)to capture and track facial expressions in the video.2.Generate face interactive animation parameters.In order to be able to generate the three-dimensional parameters that drive the animation of the digital face model,this paper proposes a method based on the SFM-3DMM three-dimensional face deformable model to estimate the depth information of the face from the two-dimensional video frame.The SFM-3DMM algorithm is used to 3D fit the 2D face feature points obtained from the tracking to obtain a pose-adaptive 3D face and output the rigid motion parameters and facial expression parameters of the face.3.Human face interactive animation synthesis scheme based on the monocular camera.Based on the existing face interaction animation synthesis scheme,this paper proposes a face interaction animation synthesis scheme based on the skeletal skin model.The skeletal skin model is a bionic model with the advantages of strong realism and easy driving.The facial animation parameters extracted from the facial motion capture are used to programmatically control the sparse bone nodes of the character's head to generate realistic facial animation sequences.