Facial expression detection method, device and system, facial expression driving method, device and system, and storage medium

A facial expression driving method, comprising:

• obtaining an image to be processed;
• inputting the image to be processed into a neural network model to obtain a facial expression parameter of a target face in the image to be processed;
• and
• driving an object to be driven based on the facial expression parameter of the target face,
• wherein the neural network model is used for fitting a face relation below:
• V=V_bar+A_id*a_id+A_exp*a_exp;
• where V represents coordinates of a first feature point on a face in an input image input into the neural network model, V_bar represents coordinates of a second feature point on an average face, the second feature point corresponds to the first feature point, a_id and a_exp respectively represent a shape parameter and a facial expression parameter of the face in the input image, A_id and A_exp respectively represent a reference shape displacement related to a shape of the face and a reference facial expression displacement related to a facial expression of the face, and V_bar, A_id and A_exp are preset values, and the input image includes the image to be processed.

A facial expression driving system, comprising a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions are capable of being executed by the processor for performing:

• obtaining an image to be processed;
• inputting the image to be processed into a neural network model to obtain a facial expression parameter of a target face in the image to be processed;
• and
• driving an object to be driven based on the facial expression parameter of the target face,
• wherein the neural network model is used for fitting a face relation below:
• V=V_bar+A_id*a_id+A_exp*a_exp;
• where V represents coordinates of a first feature point on a face in an input image input into the neural network model, V_bar represents coordinates of a second feature point on an average face, the second feature point corresponds to the first feature point, a_id and a_exp respectively represent a shape parameter and a facial expression parameter of the face in the input image, A_id and A_exp respectively represent a reference shape displacement related to a shape of the face and a reference facial expression displacement related to a facial expression of the face, and V_bar, A_id and A_exp are preset values, and the input image includes the image to be processed.