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

A collaborative shape segmentation method based on graph convolution neural network

A convolutional network and collaborative segmentation technology, applied in the field of geometric modeling and analysis of graphics, can solve the problem of high time complexity in the training process and achieve high precision

Inactive Publication Date: 2019-01-22
HANGZHOU DIANZI UNIV
View PDF4 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the disadvantage of this type of method is that firstly it also depends on the given training data set, and secondly, the time complexity of the training process for the labeled data is very high

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 collaborative shape segmentation method based on graph convolution neural network
  • A collaborative shape segmentation method based on graph convolution neural network
  • A collaborative shape segmentation method based on graph convolution neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] In conjunction with the appendix, the technical solution of the present invention is clearly and completely described through specific implementation examples.

[0021] 1. Network structure diagram

[0022] Such as figure 1 As shown, the system is mainly divided into three steps. First, the shape is over-segmented into 30 sub-slices. A graphical model is then constructed for these sub-slices through feature extraction. Finally, through the graph convolutional network model, the segmentation results of other models on the graph are learned.

[0023] 2. The model is over-segmented

[0024] figure 2 It shows that we have over-segmented the model. The system uses normalized segmentation to segment the model. We divide each model into 30 sub-slices. Their boundaries basically coincide with the feature lines. Therefore, the segmentation problem is transformed into the clustering problem of over-segmented sub-slices.

[0025] 3. Network output description

[0026] i...

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 shape collaborative segmentation method based on graph convolution network. The method of the invention comprises the following steps: a given group of shapes are overly segmented into sub-slices, and a graph model among the sub-slices is constructed. some of the sub-slices are labelled; A graph convolution network is constructed to propagate the labeled sub-chip label information to other non-labeled sub-chips. The invention applies graph convolution network to the field of shape collaborative segmentation, and the invention can obtain higher accuracy result comparedwith other methods at present.

Description

technical field [0001] The invention relates to geometric shape modeling and analysis technology of graphics, which can be widely used in three-dimensional games, modeling, simulation and other fields as a technical basis. Background technique [0002] Shape segmentation refers to dividing a shape into a limited set of sub-shapes each with simple semantics. This technology can be widely used in various fields of graphics, such as three-dimensional games, modeling, simulation, pattern recognition of models, etc. Early work mainly focused on the segmentation of a single shape, but the efficiency was low. Recently, many scholars have studied the segmentation of a group of shapes at the same time and established the correspondence between them, that is: shape collaborative segmentation. It can effectively assist in solving many shape processing problems, such as modeling, model retrieval, and texture mapping. [0003] Currently, there are unsupervised and supervised methods fo...

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): G06T7/11G06T19/20
CPCG06T19/20G06T2219/2021G06T7/11
Inventor 吴子朝秦飞巍王毅刚
Owner HANGZHOU DIANZI 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