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

Unsupervised segmentation method for terracotta army point clouds based on combination of region growing and deep learning

A deep learning and region growing technology, applied in the field of computer vision, can solve problems such as low efficiency in repairing terracotta warriors and horses

Pending Publication Date: 2021-05-28
NORTHWEST UNIV(CN)
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But even so, the efficiency of museum staff or scientific and technological workers in restoring the Terracotta Warriors is still low

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
  • Unsupervised segmentation method for terracotta army point clouds based on combination of region growing and deep learning
  • Unsupervised segmentation method for terracotta army point clouds based on combination of region growing and deep learning
  • Unsupervised segmentation method for terracotta army point clouds based on combination of region growing and deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] This embodiment provides an unsupervised segmentation method for point clouds of terracotta warriors and horses based on the combination of region growing and deep learning, which is characterized in that unsupervised semantic segmentation is performed on point cloud data, specifically including:

[0031] (1) Normal vector prediction based on the three-dimensional coordinate data of the point cloud

[0032] Calculate the nearest point for each point p in the point cloud P, and then calculate the value of the surface normal vector according to the nearest point, and judge whether the value of the normal vector is facing the viewpoint, or reverse the normal vector;

[0033] (2) Pre-segment the point cloud based on the normal vector

[0034] First, K-Nearest Neighbors is calculated for the entire point cloud to obtain the K nearest neighbors of each point. Then, a random point is selected as a starting seed and added to the set of available points to start the algorithm. F...

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 an unsupervised segmentation method for terracotta army point clouds based on combination of region growth and deep learning. The method comprises the steps of performing unsupervised semantic segmentation on point cloud data, performing normal vector prediction according to three-dimensional coordinate data of point cloud, and then performing pre-segmentation based on normal vectors on the point cloud; carrying out pre-segmentation on a point cloud by using SRG-Net, then carrying out partial segmentation on the point cloud by using SRG-Net, finally, automatically adjusting a segmentation result, and enhancing a result in the SRG-Net according to a pre-segmentation result in SRG so as to obtain a segmentation label of the point cloud. Unsupervised part segmentation of the terracotta warrior point clouds is realized, and a good effect can be obtained for segmentation of general point clouds except for the terracotta warrior point clouds.

Description

technical field [0001] The present invention relates to the field of computer vision, in particular to an unsupervised segmentation method for point clouds of terracotta warriors and horses based on the combination of region growing and deep learning, which can be used to segment three-dimensional point clouds of terracotta warriors into point cloud collections, such as: head, armor, Hands, legs. Background technique [0002] At present, there are many fragments of terracotta warriors and horses excavated from ancient tombs. The current museum staff mainly restores terracotta warriors and horses by hand. They can only complete single-digit terracotta warriors and horses in a year, which is low in efficiency. With the development of 3D equipment and the improvement of the computing power of GPU hardware equipment, it has become possible for museum staff or scientific and technological workers to digitize the Terracotta Warriors in 3D. The segmentation of 3D point cloud data ...

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 Applications(China)
IPC IPC(8): G06K9/34G06K9/62G06N3/04
CPCG06V10/267G06N3/045G06F18/23G06F18/24147
Inventor 胡尧李康周伟
Owner NORTHWEST UNIV(CN)
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