基于多粒度特征的行人跟踪检测结合算法

王子晔; 苗夺谦; 赵才荣; 罗晟; 卫志华

doi:10.7544/issn1000-1239.2020.20190280

基于多粒度特征的行人跟踪检测结合算法

王子晔¹,
苗夺谦^1,2,
赵才荣^1,2,
罗晟¹,
卫志华^1,2

¹( 同济大学计算机科学与技术系上海 201804)
²( 嵌入式系统与服务计算教育部重点实验室(同济大学) 上海 201804) (yeziwang@tongji.edu.cn)

基金项目: 国家重点研发计划项目(213)；国家自然科学基金项目(61976158,61673301)

详细信息

中图分类号: TP181
计量
- 文章访问数: 1341
- HTML全文浏览量: 4
- PDF下载量: 553
出版历程
- 发布日期: 2020-04-30

A Pedestrian Tracking Algorithm Based on Multi-Granularity Feature

¹( Department of Computer Science and Technology, Tongji University, Shanghai 201804)
²( Key Laboratory of Embedded System and Service Computing (Tongji University), Ministry of Education, Shanghai 201804)

Funds: This work was supported by the National Key Research and Development Program of China (213) and the National Natural Science Foundation of China (61976158, 61673301).

摘要

摘要: 对于一些较为流行的应用，例如视频场景监控，对行人的长期有效跟踪是应用的基础.尽管对目标检测与跟踪的相关技术研究已经有了很长的历史，但是如何实时并较为准确地实现目标行人跟踪目前仍然是一个活跃的研究领域.基于多粒度的思想，提出了一种改进的行人跟踪算法，将卷积特征与底层颜色特征结合，对基于深度学习的跟踪算法GOTURN(generic object tracking using regression networks)得到的跟踪结果进行判断决策，结合目标检测对跟踪结果进行修正.实验结果表明：与单一的跟踪算法相比，多粒度决策的跟踪算法能够更加准确地对目标行人进行跟踪，可以显著提高跟踪精度.
- 行人跟踪 /
- 目标检测 /
- 深度学习 /
- 多粒度 /
- 鲁棒性
Abstract: Recently in some popular applications, such as video scene surveillance, long-term effective pedestrian tracking is the basis of these applications. Although the related technology of target detection and target tracking have a long history, how to achieve real-time and accurate pedestrian tracking is still an active research field and needs to be solved. At present, most pedestrian tracking methods only use hand-designed features to track or only use deep learning to extract features, which are not good ways to represent the features of the target because the use of one single feature will restrict the expression of the features. Therefore, multi-granularity hierarchical features are used in this paper to achieve more stable pedestrian tracking. This paper proposes an improved pedestrian tracking algorithm. The algorithm adopts the idea of multi-granularity, combines convolutional feature with bottom color feature, makes decision on the tracking result obtained by GOTURN, a tracking algorithm based on deep learning, and modifies the tracking result with target detection. This paper uses Pascal VOC data set for model training, and uses OTB-100 and VOT 2015 data sets for testing. The experimental results show that the tracking algorithm based on multi-granularity decision can track target pedestrians more accurately than a single tracking algorithm and the tracking accuracy is improved obviously.
- pedestrian tracking /
- object detection /
- deep learning /
- multiple-granularity /
- robustness

HTML全文

参考文献(0)

施引文献(25)

期刊类型引用(10)

1.	汤书苑，周一青，李锦涛，刘畅，石晶林. 基于特征校准的双注意力遮挡行人检测器. 西安电子科技大学学报. 2024(06): 25-39 . 百度学术
2.	曹玉东，陈冬昊，曹睿，赵朗. 融合Mask R-CNN的在线多目标行人跟踪方法. 计算机工程与科学. 2023(07): 1216-1225 . 百度学术
3.	曹健，陈怡梅，李海生，蔡强. 基于深度学习的道路小目标检测综述. 计算机工程. 2023(10): 1-12 . 百度学术
4.	徐放，苗夺谦，张红云. 基于多粒度的Transformer目标检测算法. 计算机科学. 2023(11): 143-150 . 百度学术
5.	陈立，张帆，郭威，黄赟. 面向遥感图像的多阶段特征融合目标检测方法. 电子学报. 2023(12): 3520-3528 . 百度学术
6.	宋思雨，苗夺谦. 基于多粒度空间混乱的细粒度图像分类算法. 智能系统学报. 2022(01): 144-150 . 百度学术
7.	潘云磊. 基于改进蚁群算法的行人运动特征跟踪提取方法. 河北北方学院学报(自然科学版). 2022(07): 1-6 . 百度学术
8.	欧群雍，谭同德，袁红斌. 结合CNN和Bi-LSTM的多行人目标检测跟踪方法. 无线电工程. 2022(09): 1633-1641 . 百度学术
9.	袁小鑫，胡军，黄永洪. 针对目标检测器的假阳性对抗样本. 计算机研究与发展. 2022(11): 2534-2548 . 本站查看
10.	史进玲，张江维，程菊明. 多标记的不完备多粒度决策系统的最优粒度选择. 许昌学院学报. 2021(02): 103-108 . 百度学术