Face rotation model generation method and device, computer device and storage medium

Abstract

The invention discloses a face rotation model generation method and device, a computer device and a storage medium. The method comprises the following steps: acquiring a face training picture; obtaining a real position map according to the three-dimensional face parameters and a preset average face model; extracting target features in the face training picture, inputting the target features into aface rotation model, and obtaining three-dimensional position information of all second key points output by the face rotation model; setting weight values of the three-dimensional position information of each first key point and the three-dimensional position information of each second key point in a mean square error loss function through a mask; inputting the three-dimensional position information of each first key point and the three-dimensional position information of each second key point into a mean square error loss function to obtain a loss result; when the loss result is smaller than or equal to the preset loss value, confirming that training of the face rotation model is completed. The rotation angle needed for rotating the face picture into the front face picture can be accurately calculated.

Description

technical field [0001] The present invention relates to the field of model construction, in particular to a method, device, computer equipment and storage medium for generating a human face rotation model. Background technique [0002] For non-frontal faces, face recognition technology in the field of artificial intelligence is not mature enough, and face recognition technology for non-frontal faces has always been a difficult problem in the field of face recognition. The face alignment technology of non-frontal faces in the existing face recognition system usually can only solve the rotation of the roll angle in the face pose, and directly use the similarity transformation for the large-angle yaw angle or pitch angle, so, After the alignment, the face will have a large deformation, which is not conducive to the subsequent face recognition; at the same time, the face alignment of the non-frontal face in the prior art cannot obtain the accurate three-dimensional coordinate in...

AI Technical Summary

Problems solved by technology

[0003] Based on this, it is necessary to address the above-mentioned technical problems and provide a method, device, computer equipment and storage medium for generating a face rotation model, so as to solve the current problem of not being able to accurately calculate the rotation angle required to rotate a face photo into a front face photo

Images