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

Image attention visual target tracking method

A technology of target tracking and attention, which is applied in the field of visual target tracking of graph attention, can solve the problem of target features being overwhelmed, and achieve the effects of eliminating defects, simple and effective framework, and improving accuracy and speed

Pending Publication Date: 2021-08-13
ZHEJIANG UNIV OF TECH
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] Since traditional cross-correlation operations bring in a lot of background information, this can overwhelm the target features and lead to sensitivity to similar disturbances

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
  • Image attention visual target tracking method
  • Image attention visual target tracking method
  • Image attention visual target tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] In order to make the present invention easier to understand and its advantages clearer, the technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the drawings and specific embodiments.

[0032] refer to figure 1 , a graph attention visual object tracking method, comprising the following steps:

[0033] (1) Cut out the selected target tracking data set, and cut out the template image and the search area image from the original data set according to the position of the target in the image to form the training set input by the two branches of the Siamese neural network;

[0034] (2) Build a fully convolutional twin neural network for extracting image features. This network contains two convolutional neural networks with identical structures as branches, which are used to extract the cropped template image and search area image in (1). Features, use the mask to extract the features in the bounding box of the t...

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 an image attention visual target tracking method. The method comprises the following steps: (1) cutting a selected target tracking data set; (2) constructing a fully convolutional twin neural network for extracting image features; (3) building an image attention module; (4) building a classification regression network; (5) calculating a surrounding frame and a foreground score corresponding to each pixel point on the feature response graph after the feature response graph passes through the classification regression network, further calculating a total score of each pixel point according to the corresponding surrounding frame and the foreground score, and taking the point with the highest score as the central point of the tracked target; and (6) training the model by using a training data set to obtain a trained network model, and performing target tracking and positioning on the to-be-tested model by using the network. According to the invention, the tracking precision and speed are improved.

Description

technical field [0001] The method relates to the field of visual object tracking, more specifically, a graph attention visual object tracking method. Background technique [0002] For a long time, the research of visual target tracking method has been one of the hot spots in the field of computer vision, and it has played an indispensable role in scenarios such as automatic assisted driving, intelligent video surveillance, and human-computer interaction. The target tracking method in vision requires accurate and continuous prediction when predicting the scale and position of the target in consecutive frames, so as to determine the dynamic information of the target movement, such as speed, direction, etc., so as to complete more advanced visual tasks. Since the tracked target is often disturbed by a series of factors such as occlusion, deformation, and background disturbance, it is still a huge challenge to establish a high-precision and generalized target tracking model. ...

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): G06T7/246G06N3/04G06N3/08
CPCG06T7/246G06N3/04G06N3/08G06T2207/10016
Inventor 程强邵燕燕郭东岩崔滢
Owner ZHEJIANG UNIV OF TECH
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