Unsupervised target dense tracking method based on video coloring

A target tracking and unsupervised technology, applied in neural learning methods, image data processing, instruments, etc., can solve problems such as unsupervised and limited target categories, and achieve the effect of improving accuracy and enhancing capabilities

Pending Publication Date: 2022-04-26
南京创思奇科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The problem of target tracking is widely studied in computer vision, and the existing target tracking often requires a large number of image labels and the tracked target categories are limited, so the unsupervised target tracking problem has attracted extensive attention and in-depth research in the academic community.

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
  • Unsupervised target dense tracking method based on video coloring
  • Unsupervised target dense tracking method based on video coloring
  • Unsupervised target dense tracking method based on video coloring

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] The present invention will be further described below in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

[0041] A method for unsupervised target intensive tracking based on video coloring provided by the present invention, according to the following steps S1-Step S5, to obtain the target tracking model, and then apply the target tracking model to complete the tracking of the preset target object;

[0042] S1. Obtain video sample frames arranged in time order and respectively containing preset target objects;

[0043] S2. Based on the convolutional neural network and SRM module, the video sample frame is used as the input, and the feature map corresponding to the video sample frame is used as the output to construct a feature extraction network, wherein the first period of the feature extraction net...

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 unsupervised target dense tracking method based on video coloring, and the method comprises the steps: constructing a target tracking model based on video sample frames which are arranged in a time sequence and respectively comprise a preset target object, and then completing the tracking of the preset target object through employing the target tracking model. The target tracking model comprises a feature extraction network used for obtaining feature maps corresponding to video sample frames, and a dynamic adjustment module used for outputting a feature map group formed by a preset number of feature maps in all the obtained feature maps. And the target prediction module is used for predicting the position parameter and the area parameter of the preset target object in the next frame of feature map of the feature map output by the feature extraction network in real time, and the method can improve the tracking precision of the preset target object and can adapt to the tracking scene of the violent change of the preset target object.

Description

technical field [0001] The invention relates to a dense target tracking method, in particular to an unsupervised dense target tracking method based on video coloring. Background technique [0002] The problem of target tracking is widely studied in computer vision, and the existing target tracking often requires a large number of image labels and the tracked target categories are limited, so the unsupervised target tracking problem has attracted extensive attention and in-depth research in the academic community. There are four main types of solutions to the existing unsupervised target tracking problem, namely: methods based on correspondence flow, methods based on temporal cycle consistency, methods based on multi-grid prediction filters, and methods based on video coloring. The present invention is based on the method of video coloring, the general flow of this type of method is: given a certain frame in the video, traverse each pixel point p of the frame i , find the mo...

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): G06T7/246G06N3/08G06N3/04G06V10/44G06V10/82
CPCG06T7/246G06N3/084G06T2207/20081G06T2207/10016G06N3/044G06N3/045
Inventor 杜森宋爱波方效林袁庆丰杨明朱同鑫
Owner 南京创思奇科技有限公司
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