Image generation method of generative adversarial network based on dual discriminators

Abstract

The present invention relates to the technical field of image generation of a generative adversarial network, especially to an image generation method of generative adversarial network based on dual discriminators. The method comprises the steps of: obtaining text data; performing processing of the text data to obtain condition vectors; obtaining random noise; inputting the random noise and the condition vectors into a generator at the same time to obtain a generation image; inputting a real image and the generation image into a first discriminator and a second discriminator; giving a first reward value to the real image by the first discriminator, giving a second reward value to the real image by the second discriminator, giving a second reward value to the generation image by the first discriminator, and giving the first reward value to the generation image by the second discriminator; and obtaining a minimum target function according to the first reward value and the second reward value. The image generation method of the generative adversarial network based on the dual discriminators improves the diversity of the generated images and cannot consume lots of calculation resources.

Description

technical field [0001] The invention relates to the technical field of generative confrontation network image generation, in particular to an image generation method based on a dual discriminator-based generative confrontation network. Background technique [0002] With the advent of generative adversarial networks, it has become possible to generate images related to their content based on textual descriptions. Generating images related to their content according to text descriptions has a good application prospect: generating portraits of suspects to help case investigation; generating trademark images to help creators inspire more inspiration, etc. [0003] However, due to the unstable training of the current generative confrontation network, it is very dependent on the well-designed network structure and careful parameter initialization, which makes the pattern of the image generated according to the text content relatively single. Some researchers have proposed to solv...

AI Technical Summary

Problems solved by technology

[0003] However, because the current training of the generation confrontation network is very unstable, it is very dependent on the well-designed network structure and careful parameter initialization, which makes the pattern of the image generated according to the text content relatively single

Some researchers have proposed to solve this problem by improving network training methods, but these methods consume a lot of computing resources

[0004] Therefore, how to provide a new generative confrontation network structure to solve the above-mentioned problem of single mode of image generation and large consumption of computing resources has become a technical problem to be solved urgently by those skilled in the art.

Images