Neural network training method and device and image generation method and device

Abstract

The invention relates to a neural network training method and device and an image generation method and device. The neural network training method comprises the steps: inputting a first random vectorinto a generation network, and obtaining a first generation image; inputting the first generation image and a first real image into a discrimination network to obtain a first discrimination distribution and a second discrimination distribution; determining a first network loss of the discrimination network according to the first discrimination distribution, the second discrimination distribution,the first target distribution and the second target distribution; determining a second network loss of the generation network according to the first discrimination distribution and the second discrimination distribution; and according to the first network loss and the second network loss, performing adversarial training on the generation network and the discrimination network. According to the neural network training method provided by the embodiment of the invention, the discrimination network can output discrimination distribution for the input image, and the authenticity of the input imageis described in a probability distribution form, and the authenticity of the input image can be considered from multiple aspects, and information loss is reduced, and the training precision is improved.

Description

technical field [0001] The present disclosure relates to the field of computer technology, in particular to a neural network training method and device, and an image generation method and device. Background technique [0002] In related technologies, a Generative Adversarial Networks (GAN) consists of two modules, namely a discriminator network (Discriminator) and a generation network (Generator). Inspired by a zero-sum game, the two networks compete against each other to achieve the best generation effect. During the training process, the discriminator learns to distinguish between real image data and simulated images generated by the network by rewarding real targets and punishing false targets, and the generator gradually reduces the penalty of the discriminator for false targets, so that the discriminator cannot distinguish real images With the generated image, the two compete with each other and evolve, and finally achieve the effect of confusing the real. [0003] In...

AI Technical Summary

Problems solved by technology

However, the authenticity of the image is reflected in multiple dimensions, such as color, texture, scale, background, etc. Using a single scalar will cause information loss to a certain extent, and give the neural network biased guidance, resulting in unstable training and image quality. bad question

Images