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

Image reconstruction network model training method, device and electronic device

A network model, image reconstruction technology, applied in the field of image processing, can solve problems such as image deviation

Active Publication Date: 2021-06-22
BEIJING BYTEDANCE NETWORK TECH CO LTD
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, these methods cannot directly generate 3D human body images, and if the intermediate results are inaccurate, the final generated image will be greatly deviated

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
  • Image reconstruction network model training method, device and electronic device
  • Image reconstruction network model training method, device and electronic device
  • Image reconstruction network model training method, device and electronic device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Embodiments of the present disclosure are described below through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Apparently, the described embodiments are only some of the embodiments of the present disclosure, not all of them. The present disclosure can also be implemented or applied through different specific implementation modes, and various modifications or changes can be made to the details in this specification based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, in the case of no conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present disclosure, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope o...

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

Embodiments of the present disclosure disclose a training method, device and electronic equipment for an image reconstruction network model. Wherein, the training method of the image reconstruction network model includes: obtaining initial parameters of the image reconstruction network model to be trained; obtaining a sample image pair, wherein the sample image pair includes a three-dimensional image and a two-dimensional image corresponding to the three-dimensional image, Wherein the three-dimensional image is represented by a first vector; the two-dimensional image is converted into a second vector through the image reconstruction network model; an error between the second vector and the first vector is calculated; based on the The error adjusts the parameters of the image reconstruction network model. The disclosure solves the problems in the prior art by converting a two-dimensional image into a vector that can represent the vertices of a three-dimensional image and supervising the vertices of the real three-dimensional image to train a model that can reconstruct a three-dimensional image from a single two-dimensional image. Technical issues that rely on intermediate results for 3D object reconstruction.

Description

technical field [0001] The present disclosure relates to the field of image processing, in particular to a training method, device and electronic equipment for an image reconstruction network model. Background technique [0002] At present, with the continuous progress of computer technology and the development of multimedia technology, 3D reconstruction technology has become a research hotspot in the field of graphics in recent years. The indirect method refers to the reconstruction of 3D objects from one or more 2D images, including 3D object reconstruction based on statistical models, 3D object reconstruction based on multi-view geometry, 3D object reconstruction based on illumination stereo, and the rapid development in recent years based on 3D object reconstruction with machine learning. [0003] There are a variety of existing 3D human body reconstruction methods based on the SMPL (A Skinned Multi-Person Linear Model) model, such as SMPLify, which usually rely on inte...

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): G06T17/00G06N20/00
CPCG06T17/00G06N20/00
Inventor 李佩易王长虎
Owner BEIJING BYTEDANCE NETWORK TECH CO LTD