Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for generating images from Chinese text based on generative adversarial network

An image-generating, generative technology, applied in biological neural network models, image enhancement, image analysis, etc., can solve the problems of reducing the training efficiency of generative confrontation network models, and can not improve the image effect well, so as to achieve optimal generation effect , the effect of improving training efficiency

Active Publication Date: 2022-07-01
GUANGDONG POLYTECHNIC NORMAL UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method reduces the training efficiency of the generative confrontation network model due to the need to pay attention to the characteristics of the entire Chinese text describing the image, and cannot improve the effect of image generation very well.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for generating images from Chinese text based on generative adversarial network
  • Method and device for generating images from Chinese text based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0040] figure 1 Shown is a schematic flowchart of a preferred embodiment of the method for generating images from Chinese text based on generative adversarial networks provided by the present application.

[0041] like figure 1 As shown, the method includes:

[0042] S101 Perform keyword recognition on the Chinese text describing the sample image, and if at least one background keyword and at least one feature keyword are recognized, encode the bac...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The application of the present invention relates to the technical field of computer vision and natural language processing, and discloses a method and device for generating images from Chinese text based on generative adversarial networks. The method and device extract background keywords and feature keywords from Chinese text describing images to perform text semantics Embed the representation and extract the corresponding image features, and add attention mechanism processing to the image feature matrix representing the feature keyword, and then fuse the image feature matrix representing the feature keyword and the image feature matrix representing the background keyword and input it to the generation training in an adversarial network. The present application can improve the training efficiency of the generative adversarial network model, further optimize the generation effect of the important sub-regions of the feature region, and at the same time do not ignore the background features of the text, so that the generated image can be closer to the Chinese text.

Description

technical field [0001] The present application relates to the technical field of computer vision and natural language processing, and in particular, to a method and device for generating images from Chinese text based on generative adversarial networks. Background technique [0002] Chinese text generation image is to use Chinese text description to generate photo-real pictures or abstract cartoon pictures. Generating image generation from textual descriptions has become an active area of ​​research in the natural language processing and computer vision communities, with a wide range of applications including photo editing and computer-aided design. In the related art, a framework based on generative adversarial network is used to generate images from Chinese text. A generative adversarial network model consists of a generator and a discriminator. The generator is optimized to produce samples that are distributed towards the real data, thus deceiving the discriminator. Th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T11/00G06T5/50G06F40/284G06F40/30G06N3/04
CPCG06T11/00G06T5/50G06F40/284G06F40/30G06T2207/20221G06N3/045
Inventor 陈志华徐省华黄经赢魏文国
Owner GUANGDONG POLYTECHNIC NORMAL UNIV