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

Grid segmentation method based on graph convolution network

A grid segmentation and convolutional network technology, applied in image analysis, neural learning methods, biological neural network models, etc.

Pending Publication Date: 2021-04-09
ZHEJIANG UNIV
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The former often increases the amount of calculation due to the sparsity of the data, while the latter abandons the original structure of the three-dimensional object, and the amount of calculation is still relatively large

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
  • Grid segmentation method based on graph convolution network
  • Grid segmentation method based on graph convolution network
  • Grid segmentation method based on graph convolution network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The idea of ​​the present invention is: use the adjacency relationship of the faces in the grid to form a graph, use the graph convolutional neural network and feature embedding to learn features on this graph, and finally use the fully connected layer to obtain the scores belonging to each category for each face, Finally, the category to which each face belongs is predicted, which specifically includes the following steps:

[0027] Step 1: Transform the mesh model to the specified number of patches, and perform centering and scaling operations.

[0028] Step 2: Convert the model processed in step 1 into a graph representation, and perform preliminary feature extraction for each face and then input it into the trained corresponding graph convolutional neural network. For the type of part that each face in the grid belongs to Make predictions. Wherein, the graph convolutional neural network is composed of a transformation module, a graph convolution module, a feature em...

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 provides a grid segmentation method based on a graph convolution network, and the method takes the surface of a grid as a basic unit, and carries out the graph convolution operation in a dual graph formed based on the adjacency relation of the surface, so as to obtain the feature representation of the surface. In the feature acquisition stage, static and dynamic edge convolution is utilized at the same time, and the capability of learning information from potential relations between surfaces is also obtained while the actual adjacency relation is utilized. In addition, the features are further enhanced by utilizing the idea of feature embedding in instance segmentation, and finally, all parts of the grid are segmented by utilizing the enhanced features. According to the method, a better result is obtained on a data set segmented by a plurality of parts.

Description

technical field [0001] The invention belongs to the fields of computer graphics and computer vision, and in particular relates to a grid part segmentation method based on a graph convolutional network. Background technique [0002] Semantic segmentation is one of the key issues in computer vision. With the development of deep learning, semantic segmentation using neural networks in the field of two-dimensional images has been widely explored and studied. When this problem is extended to a 3D mesh, image-based operations are often not directly applicable due to its irregularity. Previous methods tend to voxelize 3D models or use multi-view 2D images to represent 3D objects, and then apply methods in 2D images. The former often increases the amount of calculation due to the sparsity of data, while the latter abandons the original structure of the three-dimensional object, and the amount of calculation is still relatively large. For the three-dimensional grid data, we transfo...

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): G06T7/11G06N3/08G06N3/04
CPCG06T7/11G06N3/08G06T2207/20081G06T2207/20084G06N3/045
Inventor 倪天宇郑友怡
Owner ZHEJIANG 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

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