Real-time object tracking method based on video inter-frame low-rank associated information consistency

A technology of associating information and video frames, applied in image data processing, instruments, calculations, etc., to achieve the effect of low calculation cost, low feature dimension, and reduced calculation amount

Inactive Publication Date: 2015-03-11
BEIHANG UNIV
View PDF3 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the problem faced by the generic model tracking method is that under the premise of a limited model library, how to design a reasonable model library description and feature description of the object to be tracked has gradually become a current research hotspot.

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
  • Real-time object tracking method based on video inter-frame low-rank associated information consistency
  • Real-time object tracking method based on video inter-frame low-rank associated information consistency
  • Real-time object tracking method based on video inter-frame low-rank associated information consistency

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0036] figure 2 The overall processing flow of the real-time object tracking method based on the consistency of low-rank correlation information between video frames is given.

[0037] This paper invents a real-time object tracking method based on the consistency of low-rank correlation information between video frames. The main steps are as follows:

[0038] 1. Target object feature description method based on compressive sensing based on local constraints

[0039] In this method, firstly, 150 random repeated observations are made on the target object through the random observation matrix R with local constraints. The random generation of observation positions is as follows: image 3 (c) shown. The observation matrix R here is a sparse matrix, whose column size is equal to the width×height of the target object, and whose row size is the nu...

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 real-time object tracking method based on video inter-frame low-rank associated information consistency. The method comprises the following steps: during the target object characteristic description period, giving position information in an initial video by a user at first and then performing characteristic description on a target object and a candidate region in a next frame by compressive sensing of local restrictions; during the inter-frame low-rank association analysis period, extracting key descriptions from a target object characteristic library to form a local characteristic library, and performing low-rank decomposition on the local characteristic library and the characteristic descriptions of the candidate region together to obtain a sparse matrix; calculating an inter-frame low-rank associated prior and cover mask; outputting a current target object position according to the sparse matrix and the inter-frame low-rank associated prior and cover mask; updating a target library, performing global low-rank analysis on a target characteristic library, and replacing the low-rank description with the best commonality with a current target object description. The method is capable of stably tracking a single object in the video for a long time in real time, and has the characteristics of high tracking speed, high tracking precision and the like.

Description

technical field [0001] The invention relates to a real-time object tracking method based on the feature description of the compressed sensing feature region in the current frame of the video and the low-rank decomposition of the low-rank correlation between the frames. Background technique [0002] Video object tracking is a very popular research direction in the field of computer vision and pattern recognition. Its specific applications include life-saving video surveillance, traffic control, and motion recognition. Although the research on video object tracking has made great achievements in recent years, some traditional limitations have not been well resolved, such as the inability to stably track high-speed moving objects with rapid object deformation and occlusion in real time. In general, current video object tracking methods can be divided into two categories: video object tracking methods based on discriminative models; video object tracking methods based on generic...

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): G06T7/20
CPCG06T7/246G06T2207/10016
Inventor 郝爱民陈程立诏李帅秦洪
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap