SiamRPN Target Tracking Based on Kalman Filter and Template Updating

doi:10.3778/j.issn.1002-8331.2202-0038

Abstract

Abstract: Aiming at the problem that SiamRPN tracking algorithm is easy to lose the tracking target when the target is moving fast and the tracking effect is affected when the template is not updated, a SiamRPN target tracking method is proposed in combination Kalman filtering and template updating. Firstly, the trained SiamRPN tracking algorithm is used to track the target, and the center point position and speed of last frame of the target object are input into Kalman filter. When the tracking frame response score obtained by the RPN network is low, the Kalman filter is used to predict the target position again, and the new tracking frame is searched. And according to the speed of the target in the last frame, the search area is adaptively expanded. Secondly, the template update network is redesigned and trained, and a channel attention mechanism is added to update the target template iteratively during tracking. Experimental results show that the success rate and accuracy rate of the proposed algorithm in OTB2015 datasets are 67.2% and 89.1% respectively, and the EAO of the proposed algorithm in VOT2016 is improved by 24.3%. Compared with other algorithms, the proposed algorithm has obvious advantages in solving target deformation and motion ambiguity problems.

Key words: target tracking, Kalman filter, template update, siamese network

摘要： 针对SiamRPN跟踪算法在目标快速运动时跟踪目标易丢失以及模板不更新影响跟踪效果问题，提出一种Kalman滤波与模板更新相结合的SiamRPN目标跟踪方法。利用训练好的SiamRPN跟踪算法对目标进行跟踪，并将上一帧目标物体的中心点位置及速度输入卡尔曼滤波器，当RPN网络得到的跟踪框响应得分较低时，利用卡尔曼滤波器重新预测目标位置，搜索得到新的跟踪框。并根据上一帧目标的速度，自适应扩大搜索区域。重新设计并训练了模板更新网络，并在其中添加了通道注意力机制，在跟踪过程中对目标模板迭代更新。实验结果表明，该算法在OTB2015的成功率和精确率分别为67.2%和89.1%，在VOT2016的EAO提升24.3%，与其他算法相比在解决目标形变和运动模糊问题具有显著优势。

关键词: 目标跟踪, Kalman滤波, 模板更新, 孪生网络

GONG Chang, SHAN Yugang, YUAN Jie. SiamRPN Target Tracking Based on Kalman Filter and Template Updating[J]. Computer Engineering and Applications, 2023, 59(8): 200-207.

巩畅, 单玉刚, 袁杰. Kalman滤波与模板更新结合的SiamRPN目标跟踪[J]. 计算机工程与应用, 2023, 59(8): 200-207.

References

[1] 单玉刚，胡卫国.尺度方向自适应视觉目标跟踪方法综述[J].计算机工程与应用，2020，56（9）：13-23.
SHAN Y G，HU W G.Review of visual object tracking algorithms of adaptive direction and scale[J].Computer Engineering and Applications，2020，56（9）：13-23.
[2] BERTINETTO L，VALMADRE J，HENRIQUES J F，et al.Fully-convolutional siamese networks for object tracking[C]//European Conference on Computer Vision，2016：850-865.
[3] LI B，YAN J，WU W，et al.High performance visual tracking with siamese region proposal network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：8971-8980.
[4] LI B，WU W，WANG Q，et al.Evolution of siamese visual tracking with very deep networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2019：16-20.
[5] GUO D，WANG J，CUI Y，et al.SiamCAR：siamese fully convolutional classification and regression for visual tracking[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：6269-6277.
[6] CHEN X，YAN B，ZHU J，et al.Transformer tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2021.
[7] 张子龙，王永雄.基于卡尔曼滤波的SiamRPN目标跟踪方法[J].智能计算机与应用，2020，10（3）：44-50.
ZHANG Z L，WANG Y X.SiamRPN target tracking method based on Kalman filter[J].Intelligent Computers and Applications，2020，10（3）：44-50.
[8] YANG T，CHAN A B.Learning dynamic memory networks for object tracking[C]//Proceedings of the European Conference on Computer Vision（ECCV），2018：152-167.
[9] LI P，CHEN B，OUYANG W，et al.Gradnet：gradient-guided network for visual object tracking[C]//Proceedings of the International Conference on Computer Vision（ICCV），2019：6162-6171.
[10] ZHU Z，WANG Q，LI B，et al.Distractoraware siamese networks for visual object tracking[C]//The 15th European Conference on Computer Vision，2018.
[11] 薛培康，杜红棉，王玮，等.基于AM5708的智能多目标跟踪监控系统设计[J].科学技术与工程，2021，21（4）：1471-1476.
XUE P K，DU H M，WANG W，et al.Design of intelligent multi-target tracking and monitoring system on AM5708[J].Science Technology and Engineering，2021，21（4）：1471-1476.
[12] FAN H，LIN L T，YANG F，et al.LaSOT：a high-quality benchmark for large-scale single object tracking[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE Press，2019：5369-5378.
[13] DANELLJAN M，BHAT G，KHAN F S，et al.ATOM：accurate tracking by overlap maximization[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），2019.
[14] DANELLJAN M，BHAT G，KHAN F S，et al.ECO：efficient convolution operators for tracking[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition（CVPR），2017.
[15] BHAT G，DANELLJAN M，GOOL L V，et al.Learning discriminative model prediction for tracking[C]//2019 IEEE/CVF International Conference on Computer Vision（ICCV），2019.
[16] DANELLJAN M，VAN GOOL L，TIMOFTE R.Probabilistic regression for visual tracking[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition，2020.
[17] VALMADRE J，BERTINETTOL，HENRIQUES J，et al.End-to-end representation learning for correlation filter based tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway：IEEE，2017：2805-2813.