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

Object three-dimensional reconstruction method, storage medium, terminal and system

A three-dimensional reconstruction and object technology, applied in neural learning methods, 3D modeling, biological neural network models, etc., to achieve the effect of improving efficiency, reducing information disturbance, and reducing workload

Active Publication Date: 2019-11-29
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the deficiency in the prior art that the three-dimensional reconstruction of an object cannot be realized based on a single picture of any angle of view, and to provide a three-dimensional reconstruction method, storage medium, terminal and system for an object

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
  • Object three-dimensional reconstruction method, storage medium, terminal and system
  • Object three-dimensional reconstruction method, storage medium, terminal and system
  • Object three-dimensional reconstruction method, storage medium, terminal and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Such as figure 1 As shown, in Embodiment 1, a method for three-dimensional reconstruction of an object specifically includes the following steps:

[0036] S01: Train the network model; in this embodiment, the generative network includes a U-shaped generative confrontation network and a 3D conditional generative confrontation network. Both the U-shaped generative confrontation network and the 3D conditional generative confrontation network are based on the generative confrontation network. Different from the original generative adversarial network. Among them, the input to the generator G in the U-shaped generative confrontation network is not a random vector, but a picture of an object from any perspective. The input to the generator G in the 3D conditional generative adversarial network is the silhouette side view of the corresponding object.

[0037] Further, training the entire network specifically includes the following sub-steps:

[0038] S011: Preprocess the da...

Embodiment 2

[0077] This embodiment is based on the same inventive concept as Embodiment 1, and provides a three-dimensional reconstruction system of an object on the basis of Embodiment 1, such as Figure 4 As shown, the system specifically includes U-shaped generative adversarial network and 3D conditional generative adversarial network. The image is input to the 3D conditional generative adversarial network; the 3D conditional generative adversarial network generates a shape mask according to the shape profile binary image features of the first fixed-view image, and then generates a 3D model of the object. Both the U-shaped generative confrontation network and the 3D conditional generative confrontation network include a generator and a discriminator, Figure 5 Schematic diagram of the 3D conditional generative confrontation network structure for shape masks, Figure 6 It is a U-shaped generative confrontation network structure diagram of the present invention.

[0078] Further, the a...

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 three-dimensional reconstruction method of an object, a storage medium, a terminal and a system, and belongs to the technical field of picture reconstruction 3D models. The method comprises: extracting high-dimensional features of a single picture at any angle, and restoring a first fixed view angle image of the object according to the high-dimensional features; and generating a shape mask according to the first fixed view angle image so as to generate a 3D model of the object. The system comprises a U-shaped generative adversarial network and a 3D conditional generative adversarial network. According to the method, the high-dimensional features of the single picture are extracted to restore the object fixed view angle, namely the first fixed view angle view, so that information disturbance can be reduced; the shape mask is generated according to the fixed view angle view, the three-dimensional reconstruction efficiency and accuracy can be improved, the methodis suitable for any view angle picture, the effect is vivid, and the requirement for reconstructing a 3D model in real time through a single object picture at any view angle is met.

Description

technical field [0001] The invention relates to the technical field of reconstructing a 3D model from a single image, in particular to a method for reconstructing a three-dimensional object, a storage medium, a terminal and a system. Background technique [0002] 3D reconstruction has a wide range of applications in the field of computer vision and modeling. In the past, researchers usually used multiple pictures from different perspectives to solve 3D reconstruction, but it is still very difficult to achieve 3D reconstruction from a single picture, because it requires a strong model understanding ability to learn from a low dimensional space to predict its shape information. [0003] Recently, researchers have made great progress in voxel-predictive 3D reconstruction using CNNs. This type of method usually considers using a fixed viewing angle or a small number of viewing angles, which is not suitable for practical applications, because in practical applications, objects ...

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): G06T17/00G06N3/08
CPCG06T17/00G06N3/08G06T2200/08
Inventor 匡平李凡何明耘彭亮
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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