Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Gray level image coloring method based on VAE-GAN and mixed density network

A VAE-GAN, grayscale image technology, applied in the field of computer vision, can solve the problems of single coloring result, the consistency of coloring structure cannot be guaranteed, etc., to improve the coloring quality, the coloring results are vivid and diverse, and the effect of suppressing color overflow

Active Publication Date: 2021-06-18
SOUTH CHINA UNIV OF TECH
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The first purpose of the present invention is to solve the problem that the coloring result in the prior art is single and the structural consistency of the coloring cannot be guaranteed, and proposes a method based on VAE-GAN (variational autoencoder-generated confrontation network) and mixed density network ( MixtureDensityNetwork, MDN) grayscale image coloring method, which can effectively improve the subjective and objective quality of image coloring

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
  • Gray level image coloring method based on VAE-GAN and mixed density network
  • Gray level image coloring method based on VAE-GAN and mixed density network
  • Gray level image coloring method based on VAE-GAN and mixed density network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] This embodiment provides a grayscale image coloring method based on VAE-GAN and mixed density network, such as figure 1 shown, including the following steps:

[0048] S1. Transform the color image in the data set from the RGB color space to the Lab color space, and obtain a grayscale image. Here, in this embodiment, the image is first scaled to a uniform size of 64×64, and then the color space conversion is performed.

[0049] S2. Construct a VAE-GAN model, make the VAE-GAN learn the ab channel color domain representation of the color image in the data set, and save the learned ab channel color domain representation.

[0050] Such as figure 2 As shown, the VAE-GAN model includes two main parts: VAE part and GAN part. Among them, the main function of VAE is to obtain the characteristics of the color gamut by reconstructing the input color gamut, while the traditional VAE model uses the reconstruction loss of the pixel-by-pixel error square in the training process, re...

Embodiment 2

[0081]This embodiment provides a grayscale image coloring device based on VAE-GAN and a mixed density network, which can implement the grayscale image coloring method in Embodiment 1. The device comprises a sequentially connected color space transformation module, a VAE-GAN model generation module, a mixed density network model generation module and a coloring module, and the VAE-GAN model generation module is also connected with the color space transformation module and the coloring module.

[0082] Wherein, the color space conversion module is used to transform the color image in the data set from the RGB color space to the Lab color space, and obtain a grayscale image;

[0083] The VAE-GAN model generation module is used to construct the VAE-GAN model, so that the VAE-GAN learns the ab channel color domain representation of the color image in the data set, and saves the learned ab channel color domain representation;

[0084] The mixed density network model generation modul...

Embodiment 3

[0088] This embodiment provides a computer-readable storage medium, which stores a program. When the program is executed by a processor, the grayscale image coloring method based on VAE-GAN and mixed density network in Embodiment 1 is implemented, specifically:

[0089] S1, transform the color image in the data set from the RGB color space to the Lab color space, and obtain a grayscale image;

[0090] S2. Construct the VAE-GAN model, make the VAE-GAN learn the ab channel color domain representation of the color image in the data set, and save the learned ab channel color domain representation;

[0091] S3. Construct a mixed density network model, use the grayscale image obtained in step S1 as model input, use the saved ab channel color domain representation as a label, and make the mixed density network model learn a mixed Gaussian distribution as a multimodal distribution of color domain representation ;

[0092] S4. For the grayscale image to be tested, first input the gray...

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 invention discloses a gray level image coloring method based on VAE-GAN and a mixed density network. The method comprises the following steps: firstly constructing a VAE-GAN model, converting a color image in a data set into a Lab color space, and obtaining a gray level image; learning ab channel color domain representation of the color image by using VAE-GAN; and secondly, constructing a mixed density network model, and learning mixed Gaussian distribution by taking a gray level image as input and ab channel color domain representation as a label. In practical application, a to-be-detected gray level image is input into the trained mixed density network model, the mixed density network model outputs a corresponding mixed distribution coefficient, a corresponding color domain representation is sampled from the mixed distribution coefficient, and then a decoder of the trained VAE-GAN model is used to decode the color domain representation to obtain a coloring result of the grayscale image. The VAE-GAN and the mixed density network are integrated, so that the image coloring quality is effectively improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to an improved grayscale image coloring method based on VAE-GAN and mixed density network. Background technique [0002] With the vigorous development of multimedia technology and digital economy, images and videos have become indispensable carriers of information dissemination in people's daily life. As an important research field of computer vision, image coloring has a wide range of applications in industrial production, medical care, education, transportation and other links. For example, in the early film and television industry, due to the limitations of equipment and technology, only black-and-white or grayscale videos or images can be shot and recorded. It is necessary to remake the coloring of the video or image; another example is in the animation production industry, the animation works are often produced by the animator first drawing the line draft image without...

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
IPC IPC(8): G06T11/40G06T5/00G06T7/90
CPCG06T11/40G06T7/90G06T2207/10024G06T5/92Y02P90/30
Inventor 王恺刘文顺
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com