Facial image conversion method based on cycle generative adversarial network

Abstract

The invention provides a facial image conversion method based on a cycle generative adversarial network. The main content of the method comprises a Wasserstein generative adversarial network (WGAN), astructural similarity (SSIM) loss, background subtraction method and face mask and a generative adversarial network (GAN). The method comprises the steps of using a generator network and a discriminator network to compete against each other, forming the cycle GAN by using a traditional GAN loss function and a new cyclic consistency loss function, improving the WGAN, improving the training of the GAN through loss, matching the SSIM loss and brightness, contrast and structural information of a generated image and an input image, inputting a binary mask and the images during training, and applying an element product to reconstruct the loss. According to the method, the cycle generative adversarial network is used, the method has higher consistency and stability in converting facial expressions, facial details and edge details can be processed well, and thus a converted image is more natural and more realistic.

Description

Technical field [0001] The present invention relates to the field of image conversion, in particular to a face image conversion method based on a cyclic generation confrontation network. Background technique [0002] As deep learning has made significant progress in computer vision problems such as image classification, object detection, and image segmentation, deep learning is considered to be able to extract high-level semantic features of images. Therefore, many interesting image applications have gradually emerged, and one of the image conversion applications that has emerged in recent years is the most popular. Image conversion is to convert one type of picture into another type of picture or exchange two facial images, referred to as "face change". The "face-changing" technology is widely used by young people today. It can convert a man’s face to a woman’s face or a woman’s face to a man’s face. It can also match a variety of different expressions to the face, and it can a...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems of insufficient precision and poor stability when processing edge details, the purpose of the present invention is to provide a facial image conversion method based on a recurrent generation confrontation network, which uses a generator network and a discriminator network to fight against each other, and utilizes traditional GAN The loss function and the new cycle consistency loss function make up the cycle GAN, then improve the WGAN, and improve the training of the GAN through its loss, then the SSIM loss matches the brightness, contrast and structure information of the generated image and the input image, and the binary The mask is input along with the image, and an element-wise product reconstruction loss is applied

Images