A method for generating frontal face images from side face images based on generative adversarial networks

Abstract

The invention discloses a method for generating a frontal face picture based on a side face picture of a generative confrontation network. The method uses the collected side face picture and the front face picture for the training of the generation confrontation network, and then collects the side face picture again and inputs it into the training In a good generative confrontation network, output a frontal face image. The method of generating the front face picture of the present invention The method of generating the front face picture of the present invention ensures the existence of local details of the face through the dual-path generator, and accelerates the extraction of effective features by the contour sub-generator by predicting the position of the sensory organs and the contour. The splicing method ensures that both features can act on the result, thereby generating high-quality images that can be used for face recognition. The method of the present invention retains local features and global features, and the generation of front face pictures has the characteristics of accuracy and high efficiency.

Description

technical field [0001] The present invention relates to the technical field of face generation, in particular to a method for generating a frontal face picture based on a side face picture of a generative confrontation network. Background technique [0002] The current face recognition system and face database use frontal face pictures as standard data. Therefore, the trained face recognition model has a good performance in the recognition of frontal faces, but it is extremely difficult for the input data of profile faces. Difference. In real life, most of the data captured by the camera is profile data, which cannot be directly applied to the face recognition system. Especially for the security field or the smart home field. For the former, a large number of cameras will be deployed in public places to search and record criminals, and criminals will deliberately avoid the cameras, so that the cameras cannot capture or only capture his profile, while the face recognition i...

AI Technical Summary

Problems solved by technology

[0005] 1. The method based on feature extraction cannot be applied to mature face recognition modules;

[0006] 2. The method based on feature generation will cause the lack of details or the invalidity of features

Images