结合双注意力与特征融合的孪生网络目标跟踪

doi:10.13190/j.jbupt.2021-169

北京邮电大学学报 ›› 2022, Vol. 45 ›› Issue (4): 116-122.doi: 10.13190/j.jbupt.2021-169

结合双注意力与特征融合的孪生网络目标跟踪

李雪, 李晓艳, 王鹏, 孙梦宇, 吕志刚

西安工业大学电子信息工程学院, 西安 710021

收稿日期:2021-08-03 出版日期:2022-08-28 发布日期:2022-06-26
通讯作者: 王鹏(1978—),男,教授,硕士生导师,邮箱:wang_peng@xatu.edu.cn。 E-mail:wang_peng@xatu.edu.cn
作者简介:李雪(1997—),女,硕士生。
基金资助:
国家自然科学基金项目（62171360）；陕西省重点研发计划项目（2022GY-110）；西安工业大学校长基金面上培育项目（XGPY200217）

Siamese Network Target Trackingthat Combines Dual Attention and Feature Fusion

LI Xue, LI Xiaoyan, WANG Peng, SUN Mengyu, Lü Zhigang

School of Electronic and Information Engineering, Xi'an Technological University, Xi'an 710021, China

Received:2021-08-03 Online:2022-08-28 Published:2022-06-26

摘要/Abstract

摘要： 基于孪生网络视觉跟踪的进化和深层网络目标跟踪算法在目标被遮挡和外观形变时的跟踪成功率不高，鲁棒性不强，对此，提出了一种结合双注意力与特征融合的孪生网络目标跟踪算法。首先，采用通道和空间注意力模块增强目标信息，抑制图像中的干扰信息，提高模型的准确度；然后，对注意力层输出的浅层和深层特征信息进行多层特征融合，得到表现力更好的目标特征，提高跟踪成功率；最后，引入在线模板更新机制，减少了跟踪漂移，提高了跟踪鲁棒性。使用OTB100测试集进行实验，实验结果表明，改进后算法的跟踪成功率比改进前算法的跟踪成功率提高了1.3%；在具有遮挡和形变属性的4个测试序列下，改进后算法的平均重叠率提高了3%，中心位置的平均误差降低了0.37个像素点，针对遮挡和外观形变时的鲁棒性更好。

关键词: 目标跟踪, 孪生网络, 双注意力模块, 特征融合, 在线模板更新

Abstract: In order to solve the problem that the evolution of siamese visual tracking with very deep networks (SiamRPN++) algorithm when the target is occluded and deformed. A siamese network target tracking algorithm combining dual attention and feature fusion is proposed. Firstly, the channel and spatial attention module are used to enhance the target information, suppress the interference information in the image and improve the accuracy of the model; then, multi-layer feature fusion is carried out for the shallow and deep feature information output from the attention layer to obtain better expressive target features and improve the tracking accuracy; finally, the online template update mechanism is introduced to reduce tracking drift and improve the tracking robustness. The OTB100 dataset is used for experimental and the results show that the success rate of the improved algorithm is increased by 1.3% compared with SiamRPN++, indicating that the tracking accuracy of the algorithm is higher; under the four test sequences with occlusion and deformation attributes, the average overlap rate of the improved algorithm is increased by 3%, and the average center position error is reduced by 0.37 pixels, which is better robustness against occlusion and appearance deformation.

Key words: target tracking, siamese network, dual attention module, feature fusion, online template update

中图分类号:

TP391

李雪, 李晓艳, 王鹏, 孙梦宇, 吕志刚. 结合双注意力与特征融合的孪生网络目标跟踪[J]. 北京邮电大学学报, 2022, 45(4): 116-122.

LI Xue, LI Xiaoyan, WANG Peng, SUN Mengyu, Lü Zhigang. Siamese Network Target Trackingthat Combines Dual Attention and Feature Fusion[J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(4): 116-122.

参考文献

[1] BERTINETTO L, VALMADRE J, HENRIQUES J F, et al. Fully-convolutional Siamese networks for object tracking[M]//Lecture Notes in Computer Science. Cham:Springer International Publishing, 2016:850-865.
[2] LI B, YAN J J, WU W, et al. High performance visual tracking with Siamese region proposal network[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE Press, 2018:8971-8980.
[3] LI B, WU W, WANG Q, et al. SiamRPN++:evolution of Siamese visual tracking with very deep networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ:IEEE Press, 2019:4277-4286.
[4] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscata-way, NJ:IEEE Press, 2016:770-778.
[5] FIAZ M, MAHMOOD A, JUNG S K. Learning soft mask based feature fusion with channel and spatial attention for robust visual object tracking[J/OL]. Sensors (Basel, Switzerland), 2020, 20(14):4021[2020-07-30]. https://www.mdpi.com/1424-8220/20/14/4021.
[6] WANG Q L, WU B G, ZHU P F, et al. ECA-net:efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ:IEEE Press, 2020:11531-11539.
[7] WOO S, PARK J, LEE J Y, et al. CBAM:convolutional block attention module[M]//Computer Vision-ECCV 2018. Cham:Springer International Publishing, 2018:3-19.
[8] LIN T Y, MAIRE M, BELONGIE S, et al. Microsoft COCO:common objects in context[M]//Computer Vision-ECCV 2014. Cham:Springer International Publishing, 2014:740-755.
[9] RUSSAKOVSKY O, DENG J, SU H, et al. ImageNet large scale visual recognition challenge[J]. International Journal of Computer Vision, 2015, 115(3):211-252.
[10] REAL E, SHLENS J, MAZZOCCHI S, et al. YouTube-BoundingBoxes:a large high-precision human-annotated data set for object detection in video[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE Press, 2017:7464-7473.
[11] WU Y, LIM J, YANG M H. Online object tracking:A benchmark[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE Press, 2013:2411-2418.

[1]	周雨昊孙哲吴晓非禹可. 基于门控特征融合的中文错别字纠正模型[J]. 北京邮电大学学报, 2023, 46(4): 91-96,122.
[2]	苏兴禹可吴晓非. 基于层次门控交互融合网络的谣言检测方法[J]. 北京邮电大学学报, 2023, 46(4): 97-102.
[3]	宋宇博高嘉振. 改进YOLOv3算法的交通多目标检测方法[J]. 北京邮电大学学报, 2022, 45(5): 103-108.
[4]	晁浩, 马庆敏, 刘永利. 利用卷积回声状态网络实现脑电情感识别[J]. 北京邮电大学学报, 2022, 45(2): 36-43.
[5]	刘逸琛, 孙华志, 马春梅, 姜丽芬, 钟长鸿. 一种基于高层特征融合的网络商品分类[J]. 北京邮电大学学报, 2020, 43(5): 98-104,117.
[6]	马素刚, 赵祥模, 侯志强, 王忠民, 孙韩林. 一种基于ResNet网络特征的视觉目标跟踪算法[J]. 北京邮电大学学报, 2020, 43(2): 129-134.
[7]	田朗, 黄平牧, 吕铁军. SA-Siam++:基于双分支孪生网络的目标跟踪算法[J]. 北京邮电大学学报, 2019, 42(6): 105-110.
[8]	刘渭滨, 邹智元, 邢薇薇. 模式分类中的特征融合方法[J]. 北京邮电大学学报, 2017, 40(4): 1-8.
[9]	任楠, 杜军平, 朱素果, 李玲慧, JangMyung Lee. 基于混合特征的运动目标跟踪方法[J]. 北京邮电大学学报, 2016, 39(6): 88-92,98.
[10]	任倩倩, 孙蓓蓓, 刘勇, 郭亚红, 金虎. 无线传感器网络动态环结构下目标信息收集方法[J]. 北京邮电大学学报, 2016, 39(1): 58-62.
[11]	刘亚辉，贾庆轩，孙汉旭，高欣. 视频序列的多特征粒子滤波跟踪方法[J]. 北京邮电大学学报, 2011, 34(s1): 14-18.
[12]	韩屏李方敏罗婷. 无线移动传感器网络的分布式目标跟踪算法[J]. 北京邮电大学学报, 2009, 32(1): 90-94.
[13]	范琳，王换招，王海. WSN中基于FCM算法的多目标跟踪数据融合[J]. 北京邮电大学学报, 2006, 29(s1): 158-162.

结合双注意力与特征融合的孪生网络目标跟踪

Siamese Network Target Trackingthat Combines Dual Attention and Feature Fusion

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价