Facial Animation System Design based on Image Processing
Xueyan Du · Advances in engineering research/Advances in Engineering Research · 2016
Using computer to make face model, establish facial expression animation and exaggerate facial image processing is a hot research topic in the current, which technology has a broad application prospect and huge economic benefits.For example, remote teaching, film and television entertainment, video games and videophone could use the technology.From the current domestic and foreign research related to the subject, there still exists many problems to study because of the complexity of the facial expression. Facial feature recognition and animation generation Facial recognitionFace recognition is one of the basic problems of computer vision, which also belongs to a branch of bio-metrics.It has been widely used in the security and identity recognition, video conference, image and video retrieval based on object, entertainment and other fields.Special position is capable of detecting human face, extract face feature to save source data for facial animation.therefore, the research of facial recognition technology before the facial animation generation algorithm is very necessary.[1]Almost all of the pattern recognition theory and technology has been applied in the field of face recognition, but some special algorithm of face recognition is also developed. Facial animation generationFacial animation generation has a variety of methods, can be roughly divided into facial animation based on 2D model generation and model based on 3D facial animation.In various methods of the model, in addition to the methods such as muscle model is unique to the 3D.Most of the methods have corresponding model of 3D and 2D version, such as morphing algorithm, grid algorithm, the radial basis function (RBF) algorithm. Texture synthesis and image fusionTexture synthesis technology has natural contact face with facial animation generation.Many of the elements can be regarded as a texture, such as facial expressions, wrinkles, skin color changes, and so on.In face recognition based on image, AAM model need to do face global texture statistics and analysis.Condensation algorithm also can build model on the current image texture sampling observation.In facial animation generation of relevant technical, texture synthesis application are more wide.It is visible that the texture synthesis is of positive significance for facial animation generation technology improvement. Facial feature points localization algorithmThe extracting and positioning of facial feature points have important applications in the man-machine interface and artificial intelligence.[2]Computer is expected to reach "human nature", change from passive interaction "machine first, people obey machine" to become a machine that can take the initiative to recognize faces, learn by watching, listening, learning the computer, can use natural language to communicate with humans.In this respect, facial features is as the main mode of video input."Look" of the robot require computer able to identify and distinguish the characteristics, this chapter focuses on several kinds of algorithms of facial feature points localization based on statistics.