Virtual expression generation method and device, electronic equipment and storage medium

Abstract

The invention relates to a virtual expression generation method and device, electronic equipment and a storage medium. The virtual expression generation method comprises the steps of acquiring three-dimensional coordinates of all first vertexes and first feature vectors of multiple edges in a first expression mesh of a target object; inputting the first feature vectors of the plurality of edges into a neural network model, and obtaining a key point data offset of a to-be-generated virtual expression and a neutral expression output by the neural network model; based on the key point data offset and the three-dimensional coordinates of the first vertexes in the first expression mesh, determining the three-dimensional coordinates of a plurality of second vertexes, the second vertexes being key points in the virtual expression; and generating a first virtual expression mesh of the target object based on the three-dimensional coordinates of the plurality of second vertexes. According to the embodiment of the invention, a plurality of virtual expression meshes can be generated by utilizing the neural network model according to the expression mesh reconstructed by the neutral expression, batch production of the Blendshapes is realized, a professional designer does not need to design the Blendshapes one by one in a time-consuming and labor-consuming manner, and the production efficiency of the Blendshapes is improved.

Description

technical field [0001] The present application relates to the field of computer technology, and in particular to a virtual expression generation method, device, electronic equipment and storage medium. Background technique [0002] In the process of customized 3D face reconstruction, professional designers are usually required to design blend shapes (Blendshapes) of different expressions of specific characters one by one for the sake of authenticity of expressions, such as closing left or right eye, opening mouth, pouting, etc. Blendshapes. Using Blendshapes as the basic expression, plus the neutral expression of the character, any expression of the character can be synthesized by superposition. [0003] The basic expressions of different characters are generally not universal, because different people have different effects when doing the same basic expressions. This requires designers to repeatedly design Blendshapes, which is time-consuming, labor-intensive, and ineffic...

AI Technical Summary

Problems solved by technology

This requires designers to repeatedly design Blendshapes, which is time-consuming, labor-intensive, and inefficient, which hinders personalized face reconstruction

Images