Face expression driving method and system based on three-dimensional face reconstruction

The invention discloses a face expression driving method and system based on three-dimensional face reconstruction, and the method comprises the following steps: obtaining an input image, and carrying out the face feature extraction of the input image; inputting the face features into a coefficient regression module to obtain face coefficients; according to the shape coefficient and the expression coefficient, performing three-dimensional reconstruction processing through a preset human face three-dimensional deformation statistical model to obtain an initial reconstructed human face, and according to the position coefficient, performing projection transformation processing on the initial reconstructed human face to obtain a three-dimensional reconstructed human face image; according to the texture coefficient, performing texture reconstruction processing through a texture statistical model to obtain a reconstructed texture image, obtaining a face texture image corresponding to the input image, and performing