Supercharge Your Innovation With Domain-Expert AI Agents!

Weight-associated target tracking algorithm based on depth map

A technology of target tracking and depth map, applied in the field of visual tracking, can solve the problem of not developing depth information to deal with occlusion, and achieve the effect of improving the effect of visual tracking

Active Publication Date: 2020-09-01
TIANJIN UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Yuan et al. incorporated the depth map into the superpixel-based object representation model, but they did not exploit the depth information to deal with the occlusion problem.

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
  • Weight-associated target tracking algorithm based on depth map
  • Weight-associated target tracking algorithm based on depth map
  • Weight-associated target tracking algorithm based on depth map

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The following is a verification of the weight-associated target tracking algorithm based on the depth map proposed by the patent. Experimental results are presented in the form of accuracy plots for the position error metric and success rate plots for the overlap metric.

[0027] To quantitatively compare these tracking algorithms, figure 1 A subjective comparison is shown. From figure 1 It can be seen that the proposed algorithm has the smallest error compared with other algorithms and can track the target more accurately. There are some algorithms that lose track of the target after encountering the occlusion problem, however the proposed algorithm can solve the occlusion problem well. In addition, the proposed algorithm can obtain the most accurate object bounding box compared with other algorithms, because it adopts a more accurate scale update mechanism and an adaptive classifier update mechanism.

[0028] figure 2 Two evaluation criteria are used to evaluate...

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 present invention relates to a weight-associated target tracking method based on a depth map, including: establishment of an occlusion detection mechanism: the first step: processing the input depth image, establishing a depth information statistical function to judge the value, if it is greater than a certain The threshold is considered to be the occlusion problem of the target; the processing of the occlusion problem: establish a spatio-temporal weight function; establish a weight association model to solve the occlusion problem; when the occlusion problem of the target occurs, the weight association model will not be updated. The weight association model is updated when the occlusion problem is encountered.

Description

technical field [0001] The present invention relates to the field of visual tracking, and more specifically, to a depth map-based target tracking algorithm. [0002] technical background [0003] Visual object tracking is a hot topic in the computer field and has a wide range of applications such as security monitoring, automatic driving and other fields. In recent years, in order to solve the difficulties of visual tracking such as partial occlusion and scale change, a large number of excellent tracking algorithms have been proposed. These tracking algorithms can be divided into methods based on generative models and discriminative models. The method based on the generative model learns the surface model of the target to find the region most similar to the tracked target as the predicted tracking result. The method based on the discriminant model regards target tracking as a binary classification problem, that is, to find the region with the largest distance from the backg...

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 Patents(China)
IPC IPC(8): G06T7/262G06K9/62
CPCG06T7/262G06T2207/20056G06T2207/10028G06T2207/10016G06F18/23213
Inventor 周圆李成浩李孜孜毛爱玲杨建兴
Owner TIANJIN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com