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

Technology for image style migration

A style and image technology, applied in the field of image style transfer technology, can solve problems such as low accuracy, difficult training, and slow algorithm speed

Pending Publication Date: 2021-06-11
CHENGDU UNIVERSITY OF TECHNOLOGY
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The GAN algorithm based on deep learning has great advantages in image transfer. The present invention provides an image style transfer system based on GAN's CycleGAN optimization algorithm to solve the problems of slow speed, difficult training and low accuracy in existing algorithms.

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
  • Technology for image style migration
  • Technology for image style migration
  • Technology for image style migration

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0017] Depend on figure 1 It can be seen that the application is completed based on the Linux CentOS-78 operating system.

[0018] CPU: The memory of the CPU is 8G, and the CPU has a powerful ALU (arithmetic operation unit), which can complete arithmetic calculations in a few clock cycles. Today's CPU can reach 64bit double precision. It only takes 1 to 3 clock cycles to perform addition and multiplication of double-precision floating-point operations. The frequency of the clock c...

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 technology for image style migration. The technology comprises an experimental environment part and an algorithm research part. According to the invention, an experimental platform is a Linux CentOS-78 operating system, and a graphics card is NVIDIA Tesla V100. The algorithm research comprises the steps of preprocessing a self-made data set and training a model. The operation process of the whole system is as follows: 1, installing a Ubuntu16.0 operating system, and configuring the version of Anacoda to be Anacoda 3, the version of python to be 3.7, the version of CUDA to be 10.0 and the version of cudnn to be 7.5; 2, sending a data set after digital vectorization to a CyclGAN optimization algorithm for training; 3, comparing and evaluating the output pictures. The main purpose of the invention is to improve the accuracy of image style conversion and promote the development of artificial intelligence.

Description

technical field [0001] The invention belongs to the aspect of image processing based on artificial intelligence, and relates to an image style transfer technology. Background technique [0002] Image style transfer is a method of applying certain features or styles of one image to another image and transforming it into a specified image style. Traditional non-parametric image style transfer methods are mainly based on the rendering of physical models and the synthesis of textures. Although these methods have achieved good results, the non-parametric image style transfer method can only extract the underlying features of the image, not the high-level abstract features. When dealing with images with complex colors and textures, the final image synthesis effect is not very good. ideal. In recent years, deep learning has developed rapidly with the wave of artificial intelligence, and it has shown amazing results in the processing of massive data. Its powerful ability to learn ...

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): G06T3/00G06N20/00
CPCG06N20/00G06T3/04
Inventor 梅琪琪王洪博祝忠明
Owner CHENGDU UNIVERSITY OF TECHNOLOGY