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

Three-dimensional model classification retrieval method based on kernel sparse representation

A kernel sparse representation and 3D model technology, applied in multimedia data retrieval, multimedia data clustering/classification, multimedia data query, etc., can solve the problem of losing the ability to classify distributed data, avoid huge sample size and improve accuracy Effect

Active Publication Date: 2019-08-02
HANGZHOU DIANZI UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, SRC will lose the ability to classify data distributed in the same direction

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
  • Three-dimensional model classification retrieval method based on kernel sparse representation
  • Three-dimensional model classification retrieval method based on kernel sparse representation
  • Three-dimensional model classification retrieval method based on kernel sparse representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] The present invention proposes a three-dimensional model classification and retrieval method based on kernel sparse representation, which will be described in detail in conjunction with relevant steps below:

[0018] Step 1. For the 3D model expressed in the form of 3D point cloud, in order to reduce the amount of subsequent calculations, avoid the disaster of dimensionality, and reduce the calculation cost, on the premise of ensuring that the main features of the 3D model are preserved, a method based on the quadratic error is used as the measurement The cost edge shrinkage algorithm performs vertex reduction on the model.

[0019] The Quadric Error Metrics (QEM) algorithm has fast calculation speed and high simplification quality. For each vertex in the model, a symmetric error matrix is ​​pre-defined, and the error of the vertex is in the form of a quadratic term. For a shrinking edge, how to calculate the position of the vertex after shrinking is very important. One...

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 three-dimensional model classification retrieval method based on kernel sparse representation. The method comprises the following steps: 1, carrying out vertex simplificationprocessing on a three-dimensional model represented in a point cloud form by using an edge shrinkage algorithm based on a secondary error as a measurement cost; 2, calculating the direction of the three-dimensional model after vertex simplification, ensuring that the same kind of models can be aligned in the direction, and obtaining three characteristic direction vectors of the three-dimensionalmodel through calculation; 3, rendering the three-dimensional model based on the three characteristic direction vectors to obtain a plurality of different rendered images in corresponding different directions; 4,for the characteristics of the three-dimensional model in different data sets, comprehensively considering data characteristics and descriptor characteristics, and selecting a characteristic descriptor to extract the characteristics of the rendered image; and 5, matching the extracted feature vectors by using an improved kernel sparse representation classifier so as to realize the classification retrieval work of the three-dimensional model. The method has certain robustness, and relatively high-efficiency and reliable superior performance is obtained.

Description

technical field [0001] The 3D model classification and retrieval method based on kernel sparse representation belongs to the field of computer vision, especially using kernel techniques to implement sparse representation classifiers in new feature spaces, and extends its applicable scenarios from 2D data to 3D data. Background technique [0002] In recent years, with the rapid development of computer technology, 3D modeling technology has become increasingly mature. Along with this upsurge, 3D models have gradually developed into the fourth-generation multimedia data type after sound, image and video. At present, the 3D models on the Internet are growing exponentially, and at the same time, they have been widely used in many fields, such as scientific research, industry and medicine. It can be seen everywhere in our life, such as playing an important role in virtual environment, computer-aided design, 3D games, film and television animation, molecular biology, geographic in...

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
IPC IPC(8): G06F16/43G06F16/45
CPCG06F16/43G06F16/45
Inventor 颜成钢温洪发孙垚棋张继勇张勇东
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