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

Non-rigid three-dimensional model classification algorithm based on adaptive sparse coding fusion

A three-dimensional model, sparse coding technology, applied in the field of non-rigid three-dimensional model classification algorithm, can solve the problem that the feature descriptor is difficult to apply to all models, the structure of the three-dimensional model is complex, the scope of application is narrow, etc., to achieve a wide range of applications, high recognition accuracy The effect of high rate and strong robustness

Active Publication Date: 2020-02-11
LIAONING NORMAL UNIVERSITY
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the structural complexity of the 3D model, the existing feature descriptors are difficult to apply to all models, and the scope of application is narrow

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
  • Non-rigid three-dimensional model classification algorithm based on adaptive sparse coding fusion
  • Non-rigid three-dimensional model classification algorithm based on adaptive sparse coding fusion
  • Non-rigid three-dimensional model classification algorithm based on adaptive sparse coding fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The non-rigid three-dimensional model classification algorithm of the self-adaptive sparse coding fusion of the present invention is as follows: figure 1 As shown, follow the steps below:

[0036] a. Select the 3D model training sample set T;

[0037] b. Let the undirected weighted graph G(V,E) represent the 3D model sample M, where V represents the vertices of the 3D model, and E represents the set of edges between the vertices of the 3D model, and extract the average geodesic distance of the 3D model sample M AGD, thermonuclear HKS and shape-diameter function SDF;

[0038] c. Use the bag-of-features model method to construct intermediate-level bag-of-features representations for the extracted average geodesic distance AGD, thermonuclear HKS, and shape-diameter function SDF, namely AGD-BoF, HKS-BoF, and SDF-BoF. Proceed as follows:

[0039] c.1 Let S={s 1 ,s 2 ,...s N}∈R p×N is the low-level geometric feature of the average geodesic distance AGD, p is the vertex...

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 adaptive sparse coding fusion non-rigid three-dimensional model classification algorithm with a wide application range, and the algorithm comprises the steps: firstly extracting the average geodesic line features, thermonuclear features and shape diameter function features of a non-rigid model, so as to construct complementary multi-feature shape description; secondly, generating feature vectors of AGD-BoF, HKS-BoF and SDF-BoF through a bag-of-features model, creating three feature dictionaries through random samples, and generating a heterogeneous multi-feature weight matrix through training; and finally, realizing effective classification of the non-rigid model by using a Sotfmax classification algorithm through adaptive fusion of the feature weight matrix andsparse optimization coding. The method is higher in recognition accuracy, is higher in robustness, and is wide in application range.

Description

technical field [0001] The invention belongs to the field of classification algorithms for non-rigid three-dimensional models, in particular to a non-rigid three-dimensional model classification algorithm for adaptive sparse coding fusion with wide application range. Background technique [0002] Due to the rapid improvement of current computer hardware processing, storage capacity and Internet bandwidth, and the continuous development and progress of acquisition technologies such as 3D scanning and procedural modeling, the wide application of 3D digital models (3D models) in various fields has been promoted. The variety of 3D models Significantly improved performance and complexity. Compared with multimedia information such as text and images, 3D models not only have a large amount of geometric information data, but also have richer geometric variability, which poses a greater challenge to the further recognition and understanding of 3D models. [0003] At present, commonl...

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): G06K9/62G06K9/46
CPCG06V10/464G06F18/28G06F18/23213G06F18/24
Inventor 韩丽刘书宁周子佳
Owner LIAONING NORMAL UNIVERSITY
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