Rotary face expression learning method based on generative adversarial network

Abstract

The invention relates to a rotary face expression learning method based on a generative adversarial network. The method comprises the following main contents: a non-coupling expression learning framework (DR-GAN) based on a generative adversarial network, improvement on face images with any attitude through a conversion model, and improvement on face synthesis of target attitudes through expression of interpolations. The method includes the following processes: providing an attitude code to a decoder, increasing attitude estimation constraint in a discriminator, separating attitude change features by feature expression learnt by DR-GAN learning in an explicit manner, taking one or more face images of a person as input, generating a unified identity feature expression, and generating any number of synthesized images of the person with different attitudes. The invention brings forward a non-coupling expression learning framework based on a generative adversarial network, which is used for rotary faces and face recognition and makes further contribution to new designs in the modeling field and to innovation solutions in the lie detection field.

Description

technical field [0001] The invention relates to the field of rotated faces, in particular to a method for learning representations of rotated faces based on generative confrontation networks. Background technique [0002] Rotating faces are widely used in arresting, polygraph detection, crowd search, modeling and other fields. Specifically, in the field of capture, when catching a fugitive, the side image of the public surveillance image is used to generate a frontal image of the fugitive through rotation, which can improve the capture efficiency. In the field of polygraph detection, due to the particularity of the polygraph detection process, the camera does not directly point to the subject. At this time, the frontal expression and posture of the person can be obtained by rotating the facial image. In the field of crowd search, the image is rotated from the side to the front image to further clarify the search target. In addition, in the field of modeling, using the side...

AI Technical Summary

Problems solved by technology

Since in daily life, it is impossible to obtain a front view of a face in many cases, and it is affected by factors such as illumination, size, and posture, and there are still certain challenges in accurately performing face rotation.

Images