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

Target tracking model updating method for simply simulating time domain regularization

A target tracking and model update technology, applied in image data processing, instrumentation, computing, etc., can solve problems such as increased difficulty and difficult application, and achieve the effect of small impact, small number of hyperparameters, and simple real-time performance.

Pending Publication Date: 2020-04-17
SHAANXI SCI TECH UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The literature [Feng Li.Learning Spatial-Temporal Regularized CorrelationFilters for Visual Tracking.In IEEE CVPR,2018] proposes a control method combining time domain regularization and space domain regularization, and obtains an analytical solution to the tracking problem, but the algorithm When the idea is used in other algorithms, the analytical solution has to be recalculated, which not only increases the difficulty in the solution process, but is even difficult to apply to some problems that cannot be solved analytically.

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
  • Target tracking model updating method for simply simulating time domain regularization
  • Target tracking model updating method for simply simulating time domain regularization
  • Target tracking model updating method for simply simulating time domain regularization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0051] see figure 1 , this embodiment specifically includes the following steps:

[0052] Step 101: input image frames to be processed;

[0053] Step 102: Perform preprocessing on the image. If the target diagonal pixel distance is greater than 100, the original image is doubled, and the size and position of the target are also doubled accordingly.

[0054] Step 103: Extend the initially given target window by a factor of 1.5 and add a cosine window. Extract features from the processed image (in this algorithm, HOG, CN and gray features are taken).

[0055] Step 104: If the frame is the first frame, proceed to step 105 to directly train the tracker model parameters and start to input the next frame image for tracking; if it is not the first frame, it means that there are already model parameters, enter step 106, and use SAMF -CA algorithm operates and obtains the best target position and scale information.

[0056] Step 107: Calculate the APCE response value V of the curre...

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 the technical field of computer vision target tracking, and discloses a target tracking model updating method for simply simulating time domain regularization, which adopts asimulation time domain regularization method and utilizes the change degree of model parameters trained by a current frame and a previous frame to determine an updating rate. The method not only achieves self-adaptive updating of the tracking model, but also overcomes the limitation that the STRCF algorithm must have an analytical solution. The invention can be used for any tracking method withouta proper updating strategy. Meanwhile, the number of hyper-parameters is small, the hyper-parameters can be conveniently integrated into other tracking algorithms based on the appearance model, modelparameters are reasonably updated, model drifting is reduced, and the precision of a detection result is further improved.

Description

technical field [0001] The invention relates to the technical field of computer vision target tracking, in particular to a simple analog time domain regularization target tracking model updating method. Background technique [0002] Object tracking is a challenging research hotspot in the field of computer vision, which has a wide range of applications, such as autonomous driving, mobile robots, video surveillance, abnormal behavior analysis, and so on. In recent years, correlation filters (Correlation Filters, referred to as CF) have been introduced into the framework of target tracking, and have achieved remarkable results in both accuracy and speed. In 2010, Bolme et al. proposed a new type of correlation filter, MOSSE (Minimum Output Sum of Squared Error), which applied CF to the tracking algorithm for the first time. MOSSE uses correlation filters to model the appearance of targets and performs operations in the frequency domain, which significantly improves the tracki...

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): G06T7/246G06T7/269G06T7/40
CPCG06T7/269G06T7/40G06T7/246
Inventor 尹向雷马晓虹
Owner SHAANXI SCI TECH 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