Image processing method and device, computer readable storage medium and electronic device

Abstract

The invention provides an image processing method and device, a computer readable storage medium and an electronic device. The method comprises steps of constructing a first generative adversarial network and a second generative adversarial network, wherein the first generative adversarial network is associated with the second generative adversarial network. A first generator in the first generative adversarial network is used for generating a first image according to a first original image. A second generator in the second generative adversarial network is used for generating a second image according to a second original image. An image change loss function is constructed according to the first original image, the first image, the second original image and the second image. A loss function is constructed according the image change loss function. The first generative adversarial network and the second generative adversarial network are trained until the value of the loss function reaches a target value. Image processing is performed by use of the trained first generative adversarial network and the trained second generative adversarial network. According to the invention, transferring of image local features can be optimized; over-fitting is inhibited; and robustness is improved.

Description

technical field [0001] The present disclosure relates to the field of computers, and in particular to an image processing method, an image processing device, a computer-readable storage medium, and electronic equipment. Background technique [0002] In recent years, with the rise of artificial intelligence, deep learning has received extensive attention. At present, machine learning models are roughly divided into two types: Generative Model and Discriminative Model. Among them, the discriminant model requires an input variable x, and the output variable y is predicted by solving the conditional probability distribution p(y|x) through the discriminant model; the generative model is given some hidden information to randomly generate observation data. [0003] Generative confrontation network is a kind of generative model, including generator network (Generator Network) and discriminator network (Discriminator Network), generator network and discriminator network compete with...

AI Technical Summary

Problems solved by technology

Due to the poor effect of image local feature transfer in the prior art, while realizing local feature transfer, the ...

Images