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

A Projective Reconstruction Method Based on Trajectory Basis

A trajectory-based and projective technology, applied in the field of computer vision research, can solve problems such as large errors, achieve the effects of small errors, increase computing speed, and reduce unknowns

Inactive Publication Date: 2018-10-23
SHAANXI NORMAL UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, Akhtert's method is based on the orthographic projection model. When the depth of field of the object cannot be ignored compared with the distance from the camera to the object, the error is 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
  • A Projective Reconstruction Method Based on Trajectory Basis
  • A Projective Reconstruction Method Based on Trajectory Basis
  • A Projective Reconstruction Method Based on Trajectory Basis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] This article uses the dance video sequence in the Cameron University laboratory and converts it into an image sequence for experiments. In order to reflect the movement situation, when selecting feature points, select the arm and lower leg parts that can reflect the movement situation.

[0047] Carry out projective reconstruction according to the method of the present invention, the flow chart is as follows figure 1 As shown, the specific steps are as follows:

[0048] (1) In the image sequence (such as figure 2 and 5 ) to extract the feature point data that can reflect the motion trajectory in each image, such as image 3 and 6 shown, specifically expressed as F and P are the number of images and the number of feature points, respectively;

[0049] (2) Solve the depth factor of the image according to the feature point data, and complete the projective reconstruction. The specific implementation steps are:

[0050] (2.1) Assuming that the camera model is a pers...

Embodiment 2

[0081] This embodiment is aimed at the reconstruction of image noise changes, the basic operation steps are the same as in Embodiment 1, and the number of images F=80 and the number of feature points P=100 are generated for experimentation, the number of iterations k is 30, and ξ is 10 -4 , and add Gaussian noise with zero mean and variance from 0 to 2 in the image, repeat the operation 60 times under the condition that the primitive r is 2, 4, 6, 8 and 10, and then calculate the average value, the experiment The result is as Figure 8 shown.

[0082] From Figure 8 It can be seen that v 3r+2 The value of increases linearly with the image noise, indicating that the method is more robust. At the same time, it can also be seen that the larger the number of primitives, the greater the v 3r+2 The smaller the value, the reason is that when the matrix is ​​decomposed by singular value, the V values ​​are arranged from large to small, so the corresponding value is smaller.

[008...

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 relates to a projective reconstruction method based on a trajectory basis. Assuming that a non-rigid body is composed of multiple trajectory bases. Firstly feature point data of image sequences are extracted and feature point three-dimensional homogeneous trajectory coordinates are established. Then singular value decomposition is performed by utilizing the low-rank characteristic of an image matrix. A depth factor is solved by using row vector constraints and column vector constraints simultaneously so that projective reconstruction can be realized. According to the method, operation speed and robustness of the algorithm are enhanced, convergence speed is ensured to be fast, the re-projection effect is great, and the reconstruction effect is enabled to be closer to the actual situation and error is lower by aiming at a prospective model through combination of the computing method.

Description

technical field [0001] The invention belongs to the technical field of computer vision research, in particular to a trajectory-based projective reconstruction method for non-rigid bodies. Background technique [0002] 3D reconstruction based on image sequence is a hot issue in computer vision research, and projective reconstruction is a necessary process of 3D reconstruction, and its accuracy will directly affect the result of 3D reconstruction. 3D reconstruction has broad application prospects in the fields of biomedicine, game manufacturing, and animation production, so its research has important research significance and practical value. After nearly 20 years of development, the research on rigid bodies has approached maturity, which laid the foundation for the later research on non-rigid bodies. Most of the movements in life are flexible and belong to non-rigid bodies, but the structure will change when non-rigid bodies move, and the movement situation is complicated. ...

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 Patents(China)
IPC IPC(8): G06T7/536G06T17/00
CPCG06T17/00G06T2207/10016
Inventor 刘侍刚李丹丹彭亚丽裘国永
Owner SHAANXI NORMAL 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