Patent · US Active

3D reconstruction method based on deep learning

US11450066B2 · kind B2 · utility

1Cited by

0References

6Claims

0Family size

Assignee

BEIJING UNIVERSITY OF TECHNOLOGY · CN

Inventors

Dehui KONG · Longbeilingcun, CN
Caixia Liu · Beijing, CN
Shaofan Wang · Beijing, CN
Jinghua Li · Zhaoyuan, CN
Lichun Wang · Sichuan, CN

Key dates

Filing date	Mar 4, 2020
Grant date	Sep 20, 2022
Priority date	—
Expiry date	Jul 15, 2040

Classification

Technology area (CPC Y)Emerging Cross-Sectional Technologies
CPC primaryY02T10/40
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

3D reconstruction method based on deep learning includes the following steps: (1) The potential vector constrained in the input image is used to reconstruct the complete 3D shape of the target, and the mapping between the part and the complete 3D shape is learned, then the 3D reconstruction of a single depth image is realized. (2) Learn the intermediate feature representation between the 3D real object and the reconstructed object to obtain the target potential variables in step (1). (3) The voxel floating value predicted in step (1) is transformed into binary value by using the limit learning machine to complete high-precision reconstruction.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.