Three-dimensional face reconstruction method and device, electronic equipment and storage medium

Abstract

The invention relates to a three-dimensional face reconstruction method and device, electronic equipment and a storage medium. The three-dimensional face reconstruction method comprises the following steps: acquiring a face image to be reconstructed; preprocessing the face image to obtain a face image with a standard size; inputting the face image with the standard size into a trained face parameter recognition model to obtain a face parameter corresponding to the face image; wherein the face parameters comprise a posture parameter, a shape parameter and an expression parameter; according to the face parameters, three-dimensional vertex coordinates corresponding to the face image are obtained through calculation; and storing the three-dimensional vertex coordinates to obtain a complete file containing the three-dimensional face model. According to the three-dimensional face reconstruction method provided by the invention, three-dimensional face reconstruction based on a single face image can be realized, and high-precision three-dimensional face data does not need to be acquired.

Description

Technical field [0001] The present invention relates to the field of computer visual technology, and in particular, there is a three-dimensional face reconstruction method, a device, an electronic device, and a storage medium. Background technique [0002] Three-dimensional face rebuild has been a hot topic in computer visual fields. The three-dimensional face reconstruction can be applied to face recognition, face alignment, face animation, etc. In recent years, although researchers have proposed a number of three-dimensional face rebuild, it is still a challenge from the four-dimensional image. To simplify problems, researchers will use the three-dimensional face model or template to perform reconstruction tasks, the most common model is a three-dimensional modified model (3DMorphable Model, 3DMM). [0003] At present, 3DMM-based three-dimensional face reconstruction methods can be roughly divided into two categories: optimized methods and methods based on deep learning. [000...

AI Technical Summary

Problems solved by technology

[0005] The disadvantage of the optimization-based method is that it needs to rely on a large-scale 3D face database to construct a face model, and involves a large number of iterative fitting calculations.

The common problems of deep learning-based methods are: (1) The introduction of deep convolutional networks (such as VGG16) will increase the computational cost, increase the running time, and rely on high-performance processors

If the number of layers of the neural network is directly reduced, it is easy to cause the model to not fit well

(2) Insufficient stability and robustness of reconstruction results

Specifically, most 3D face reconstruction methods usually do not support face images with large poses (e.g., face yaw angle up to 90°), which will cause distortion of the reconstructed model

Images