Three-dimensional face construction method and device and electronic equipment

Abstract

The invention provides a three-dimensional face construction method and device and electronic equipment. The method comprises the following steps: acquiring a face image; determining a face feature vector and a 3DMM coefficient of the face image; inputting the 3DMM coefficient into a 3DMM model, and outputting a first face texture and a three-dimensional face shape of the face image; inputting thefirst face texture and the face feature vector into a texture optimization model, and outputting a second face texture; and constructing a target three-dimensional face of the face image based on thethree-dimensional face shape and the second face texture. The first face texture and the face feature vector are input into the texture optimization model, the rough first face texture is optimized,the second face texture has a finer texture structure, and therefore the target three-dimensional face constructed based on the second face texture has higher trueness, and a large-scale data set containing high-definition face textures does not need to be collected for training in a three-dimensional face construction process, so that money and human resources are saved.

Description

technical field [0001] The present invention relates to the technical field of computer vision, in particular to a method, device and electronic equipment for constructing a three-dimensional human face. Background technique [0002] In the field of computer vision, reconstructing the shape and texture of a 3D face from a single face image is an important research topic. The method based on 3DMM (3D Morphable Model, 3D face variability model) can successfully reconstruct a 3D face from a single face picture. Generally speaking, this method needs to use hundreds of face scan models based on PCA (Principal Components Analysis , Principal Component Analysis) method to construct static face shape and texture models. [0003] However, although the methods based on 3DMM and its variants can reconstruct the three-dimensional shape and texture information of the face from a single face picture, the authenticity of the reconstructed face texture is not high enough. The main reason ...

AI Technical Summary

Problems solved by technology

[0003] However, although the method based on 3DMM and its variants can reconstruct the 3D shape and texture information of the face from a single face picture, the authenticity of the reconstructed face texture is not high enough.

The main reason is that the 3DMM method performs modeling on a small amount of 3D face data collected in a laboratory environment, and this method does not perform well in uncontrolled conditions such as complex lighting conditions

If you want to reconstruct high-definition face texture information under complex lighting conditions, you need to collect a large-scale dataset containing high-definition face textures. Collecting such a dataset requires a lot of money and manpower, which is difficult for ordinary people. accomplish

Images