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

A cascaded network 3D reconstruction method and system with adaptive viewing angle

A technology of three-dimensional reconstruction and cascaded network, applied in the field of computer vision, can solve the problems of high storage cost and long time-consuming multi-view reconstruction, and achieve the effect of improving robustness, eliminating the influence of illumination and occlusion, and reducing sensitivity

Active Publication Date: 2022-06-03
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In order to solve the technical problems that lighting will have a great impact on 3D reconstruction, multi-view reconstruction takes a long time and storage costs are high, the present invention proposes a cascaded network 3D reconstruction method and system with adaptive view angles

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
  • A cascaded network 3D reconstruction method and system with adaptive viewing angle
  • A cascaded network 3D reconstruction method and system with adaptive viewing angle
  • A cascaded network 3D reconstruction method and system with adaptive viewing angle

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention will now be described with reference to the accompanying drawings.

[0032] The cascaded network three-dimensional reconstruction system with adaptive viewing angle in the embodiment of the present invention includes a camera assembly, a signal transmission line, a memory, a processor, and a program stored in the memory for rapid reconstruction by the processor. The camera assembly is composed of a detachable slide rail, a slide rail platform car, a six-axis mechanical arm, a pair of high-definition cameras with adjustable relative positions, a controller, a high-precision position sensor and a signal transmission device. Connected to the working end of the robotic arm. Due to the large scale of the neural network and high requirements for processors and memory, a mobile workstation with a high-performance graph...

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 present invention proposes a three-dimensional reconstruction method and system of a cascaded network with adaptive viewing angles. The reconstruction method mainly includes system initialization, and the camera assembly automatically collects one or more sets of images from different viewing angles for initial reconstruction. The initially reconstructed point cloud and Expand the set of existing viewing angles, estimate the viewing angle that is most helpful to improve the integrity of the reconstruction after adding it, control the camera assembly to shoot, add the existing image set, and combine it with the adjacent image set for depth estimation. The estimated result is converted into a point cloud and added to the current There are candidate point cloud sets, repeat this step until the requirements are met, and finally filter and suppress based on the point-to-point distance and statistical characteristics of the point cloud. The method described is mainly used for dense reconstruction of 3D point clouds of still objects, aiming to balance the contradiction between reconstruction integrity and calculation and storage costs, while avoiding damage to high-value, difficult-to-move objects during the reconstruction process.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a cascade network three-dimensional reconstruction method and system for adaptive viewing angle. Background technique [0002] 3D reconstruction refers to the technology of recovering the three-dimensional information of objects or scenes by using two-dimensional projection, and it is the key technology to reproduce the objective physical world in the computer virtual world. [0003] According to whether the sensor actively sends a signal to the measured object, 3D reconstruction methods can be divided into active and passive methods. The active method uses the principles of encoded structured light and time of flight (ToF) for reconstruction. The price is still relatively high, and the usage scenarios are also relatively limited; passive 3D reconstruction has low hardware requirements, and can be reconstructed using RGB images collected by ordinary cameras, mainly based ...

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): G06T17/20
CPCG06T17/20
Inventor 王好谦刘智方
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
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