Image conversion network training method, device, computer equipment and storage medium
A technology for computer equipment and image conversion, applied in computer parts, computing, instruments, etc., can solve the problems of complex implementation and poor versatility, and achieve the effect of good versatility
Active Publication Date: 2021-06-04
XIAMEN MEITUZHIJIA TECH
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Although the staged training method is effective, it is more complicated to implement because it relies on changes to the network structure of the previous stage, and its versatility is not very good.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all of them. The components of the embodiments of the application generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application.
[0040] It should ...
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
Login to View More Abstract
The present application provides an image conversion network training method, device, computer equipment, and storage medium. The image conversion network training method includes: firstly, training the first network through a downsampled image of the first image to obtain a second image; The image trains the second network to obtain a third image that approximates the upsampled image content of the second image. Determine whether the difference between the characteristic information of the third image and the reference image is lower than the threshold through the discriminant network, if not lower than the threshold, feed back the difference to the second network to adjust the network parameters of the second network, and repeat the above steps until The feature information difference between the image output by the second network and the reference image is lower than the threshold. By constructing a brand-new image conversion network to process high-resolution images, it does not depend on the previous network and has good versatility. It also solves the problem that the existing image conversion methods with good generality cannot process high-resolution images. Technical issues with making the conversion.
Description
technical field [0001] The present application relates to the field of image conversion, in particular, to an image conversion network training method, device, computer equipment and storage medium. Background technique [0002] In recent years, with the development of deep learning, many image conversion methods have emerged in the field of image conversion, especially represented by pix2pix and cyclegan series methods, among which pix2pix is a supervised learning method with paired data, and cyclegan is an unsupervised learning method without paired data. method. [0003] However, although the above methods have good versatility, they are only suitable for low-resolution image conversion. For example, pix2pix and cyclegan can only process images with a resolution of 256x256 at most, and the effect of processing images at high resolutions is often not good. good. [0004] To this end, the academic community has proposed many improvement methods, the most important of wh...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
CPCG06F18/214
Inventor 黄凯翔吴善思源王晓晶洪炜冬张伟
Owner XIAMEN MEITUZHIJIA TECH