Long-time video tracking method based on adaptive correlation filtering

A technology of correlation filtering and video tracking, applied in the field of video tracking, which can solve problems such as occlusion, template drift tracking, failure, etc.

Active Publication Date: 2019-11-19
JIANGNAN UNIV
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the current problem that the target is fast moving or severely occluded, and the template drift is likely to cause tracking failure when the target is tracked for a long time, and the robustness of the algorithm is not high, the present invention provides a long-term video based on adaptive correlation filtering tracking method

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
  • Long-time video tracking method based on adaptive correlation filtering
  • Long-time video tracking method based on adaptive correlation filtering
  • Long-time video tracking method based on adaptive correlation filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0086] This embodiment provides a long-term video tracking method based on adaptive correlation filtering, see figure 1 , the method includes:

[0087] Step 1. Obtain the first frame of the target, manually determine the center position of the target and the length and width of the bounding box (lx 1 ,ly 1 ,w 1 , g1 ), where lx 1 is the abscissa of the center position of the target, ly 1 is the ordinate of the center position of the target, w 1 is the bounding box width, g 1 is the bounding box height;

[0088] Step 2. According to the center position of the object in the first frame, dense sampling is performed with a step size of 1 pixel; when the overlap ratio of the sample and the object bounding box is greater than 0.9, the sample is assigned a positive label, and when it is less than 0.5, the sample is assigned a negative label. Generate a sample set to train the support vector machine;

[0089] What needs to be explained is that in order to automatically obtain ...

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 long-time video tracking method based on adaptive correlation filtering, and belongs to the technical field of video tracking. According to the method, HOG features, CN features and gray features are fused, feature discrimination is enhanced, meanwhile, detection suggestions are generated in combination with Edgebox, and the optimal suggestion is found to achieve self-adaption of the tracker scale and the aspect ratio; the template is prevented from being damaged by using a high-confidence tracking result, the moving speed of the target and the number of edge groups are combined to form a new adaptive updating rate, and the scale of each frame of target frame is corrected; and in the case of tracking failure, applying an incremental learning detector to recover the target position in a sliding window manner. According to the method, the correlation filter has better scale and length-width ratio adaptability, the stability of the template is better due to scalecorrection and a high confidence updating mechanism, and the method is excellent in performance under the conditions of disordered backgrounds, shielding, rapid movement of targets and the like and suitable for long-time tracking.

Description

technical field [0001] The invention relates to a long-term video tracking method based on adaptive correlation filtering, and belongs to the technical field of video tracking. Background technique [0002] Target tracking usually refers to given the initial position of the target in the first frame, estimating the position and shape of the tracking target in the subsequent video sequence, and obtaining information such as the moving direction and trajectory of the target. Nowadays, target tracking plays a pivotal role in computer vision, and has a very wide range of applications in human-computer interaction, national defense security, smart home and other fields. [0003] Target tracking is mainly divided into generative methods and discriminative methods according to different model categories. Among them, the discriminative method uses the target as a positive sample and the background area as a negative sample, and distinguishes the two significantly, and finds the tar...

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): G06K9/00
CPCG06V20/42G06V20/48G06V2201/07
Inventor 葛洪伟肖逸清杨金龙羊洁明江明
Owner JIANGNAN 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