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

A target tracking method based on context correlation and a discriminant correlation filter

A correlation filter and target tracking technology, applied in the field of visual tracking, can solve the problems of low tracking accuracy and instability, and achieve the effect of improving tracking accuracy and good robustness

Inactive Publication Date: 2019-03-29
NANJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a target tracking method based on context correlation and discriminant correlation filters, which uses feature maps and Context information training learns context-related filters and scale-related filters. The specific technical solutions are as follows:

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 target tracking method based on context correlation and a discriminant correlation filter
  • A target tracking method based on context correlation and a discriminant correlation filter
  • A target tracking method based on context correlation and a discriminant correlation filter

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] combine Figure 1 ~ Figure 3 , in an embodiment of the present invention, the present invention provides a target tracking method based on context correlation and discriminant correlation filters, the method comprising steps:

[0041] S1. Construct an end-to-end tracking network based on the correlation filter, and use the tracking network as a reference network to build a tracker for tracking and tracking objects; specifically, in the embodiment of the present invention, learn linearity by using the feature map of the training image Template, and use the same feature map to search for similar images in the test set through cross-correlation, so as to realize the construction operation of the tracking network.

[0042] Preferably, the reference network constructed is an asymmetric Siamese network; and, the method of the present invention uses the formula Define the correlation filter ω ∈ R n , obtained by feature map training; and the differential of each dependent v...

Embodiment 2

[0052] Based on the target tracking method based on context correlation and discriminant correlation filters described in Embodiment 1, the method of the present invention is specifically described through experiments; specifically, this embodiment specifically uses the VOT2017 data set, and the VOT2017 video sequence data set is used as a popular A tracking benchmark with all ground truth annotations and visual attributes, including camera_motion, empty, illum_change, motion_change, occlusion, size_change, mean, weightedmean, and pooled; preferably, this embodiment uses accuracy, robustness, and expected These three evaluation criteria are averagely overlapped to evaluate the performance of the proposed method, and the details of these three evaluation criteria are described as follows:

[0053] Accuracy: An error is detected when the bounding box predicted by the tracker has zero overlap with the ground truth. Under certain conditions, accuracy is affected by the average ove...

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 target tracking method based on context correlation and a discriminant correlation filter. The method comprises the following steps: S1, constructing an end-to-end tracking network based on correlation filter, and constructing a tracker for tracking an object with the tracking network as a reference networkS2, generating a feature map by using the first three convolutionlayers of the VGG16 model, and training and learning a context-related filter and a scale-related filter based on the feature map and the context information; 3, training a translation filter in combination with that context-related filt and the feature map, and locating a position of a tracking object with the translation filter; S4, using the scale correlation filter to calculate the proportionof the tracking object based on the position of the tracking object, and locating the position of the tracking object in the next frame in combination with the translation filter, the scale correlation filter, the feature map and the context information; The invention can effectively improve the accuracy and robustness of target tracking.

Description

technical field [0001] The invention belongs to the technical field of visual tracking, and in particular relates to a target tracking method based on context correlation and discrimination correlation filters. Background technique [0002] Object tracking is a hot issue in computer vision and has a wide range of applications, such as human-computer interaction, robotics, autonomous driving, intelligent traffic control, etc. At present, the existing target tracking methods can be mainly divided into discriminative methods and generative methods. [0003] The generative target tracking method mainly uses the generative model to describe the apparent characteristics of the moving target, that is, only pays attention to the description of the target itself and ignores the background information. The core is to study how to use the sparse representation, particle filter, mean-shift, etc. The target model is used to search for candidate targets to minimize the reconstruction err...

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): G06T7/246
CPCG06T7/248G06T2207/10016G06T2207/10024G06T2207/20024G06T2207/20081G06T2207/20084
Inventor 孙鹏朱松豪朱静怡郭文波
Owner NANJING UNIV OF POSTS & TELECOMM
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