Correlation filter tracking algorithm based on double model

ZHANG Hong-ying; WANG Hui-san; HU Wen-bo

doi:null

您当前的位置：

首页 >

文章列表页 >

Correlation filter tracking algorithm based on double model

更新时间：2020-08-13

- Correlation filter tracking algorithm based on double model
- Optics and Precision Engineering Vol. 27, Issue 11, Pages: 2450-2458(2019)
- 作者机构：
  
  中国民航大学电子信息与自动化学院天津,300300
- 作者简介：
- 基金信息：
- DOI：
  CLC： TN713
- Received：19 April 2019，
  
  Revised：30 June 2019，
  
  Published Online：15 November 2019，
  
  Published：15 November 2019
- 稿件说明：
移动端阅览
张红颖王汇三胡文博,. 基于双模型的相关滤波跟踪算法[J]. 光学精密工程, 2019,27(11): 2450-2458

ZHANG Hong-ying, WANG Hui-san,HU Wen-bo. Correlation filter tracking algorithm based on double model[J]. Editorial Office of Optics and Precision Engineering, 2019,27(11): 2450-2458
张红颖王汇三胡文博,. 基于双模型的相关滤波跟踪算法[J]. 光学精密工程, 2019,27(11): 2450-2458 DOI：

ZHANG Hong-ying, WANG Hui-san,HU Wen-bo. Correlation filter tracking algorithm based on double model[J]. Editorial Office of Optics and Precision Engineering, 2019,27(11): 2450-2458 DOI：

摘要

为解决因边界效应导致相关滤波跟踪算法不够稳健及其不能适应尺度变化的问题，提出了一种基于双模型的相关滤波跟踪算法。方法：将目标跟踪分为位置预测和尺度预测两部分，在位置滤波器模型进行位置预测阶段，先通过对待测样本进行样本增强处理，使得到的样本更符合实际场景。再通过交替方向乘子法进行位置滤波器的迭代求解，最后得到估计的目标位置。在尺度滤波器模型进行尺度预测阶段，通过在估计的目标位置处构建多尺度金字塔来训练尺度滤波器，再求解得到目标的尺度，将双模型得到的结果作为最终的跟踪结果。最后通过引入一个遮挡判据来判断是否更新模型以提高算法的鲁棒性。结果：实验表明，改进算法和经典的相关滤波跟踪算法相比，在跟踪成功率上提高了18%，在跟踪精度上提高了11%。结论：在目标被遮挡、自身尺度变化时，改进算法仍能稳定跟踪。

Abstract

For the correlation filtering tracking algorithm is not robust enough and cannot adapt to scale changes due to the boundary effect

an improved correlation filtering tracking algorithm based on double model is proposed. Methods: The target tracking consists of position prediction and scale prediction. In the stage of the position prediction

the samples were enhanced to make the samples more consistent with the actual scene.Then the solution is obtained by using the alternating?direction?method?of?multipliers(ADMM)

and the estimated target position is achieved. For scale prediction

a multi-scale pyramid was constructed in order to train the scale filter

and then the target scale is acquired. The final tracking result is decided by target position and scale together. Finally

an occlusion criterion is introduced to determine whether the model is updated or not. Results: Comparing with the classical correlation filtering tracking algorithm

the proposed algorithm boosts the tracking success rate by 18% and tracking accuracy by 11%. Conclusion: Our algorithm in this paper could track the target stably even when the target is occluded and its scale changes.

关键词

Keywords

references

Views

180

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Correlation filter tracking algorithm based on double model

A Transformer-based visual tracker via knowledge distillation

Special attribute-based cross-modal interactive fusion network for RGBT tracking

Salient region weighted correlation filter for object tracking

Siamese network tracking with redetection mechanism

Related Author

Hong-ying ZHANG

Hui-san WANG

Wen-bo HU

JIA Xingxuan

LIU Mengqiao

LI Na

PAN Jinting

HUANG Kai

Related Institution

School of Communication and Information Engineering， Xi’an University of Posts and Telecommunications

College of Electrical and Control Engineering， Xi’an University of Science and Technology

University of Chinese Academy of Sciences

Changchun Institute of Optics， Fine Mechanics and Physics Chinese Academy of Sciences

Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰