基于多模型蒸馏的时间正则化相关滤波跟踪算法

doi:10.12305/j.issn.1001-506X.2022.08.08

系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (8): 2448-2456.doi: 10.12305/j.issn.1001-506X.2022.08.08

基于多模型蒸馏的时间正则化相关滤波跟踪算法

仇祝令¹, 查宇飞^2,^3,*, 李振宇¹, 李禹铭¹, 张鹏¹, 朱川¹

1. 中国人民解放军63787部队, 新疆石河子 832099
2. 西北工业大学计算机学院, 陕西西安 710072
3. 西北工业大学宁波研究院, 浙江宁波 315000

收稿日期:2021-08-24 出版日期:2022-08-01 发布日期:2022-08-24
通讯作者: 查宇飞
作者简介:仇祝令 (1995—), 男, 硕士研究生, 主要研究方向为目标跟踪、机器学习以及计算机视觉|查宇飞 (1979—), 男, 副教授, 博士研究生导师, 主要研究方向为机器学习、计算机视觉、目标跟踪、目标检测、人工智能|李振宇 (1990—), 男, 硕士, 主要研究方向为目标跟踪、信号处理|李禹铭 (1996—), 男, 本科, 主要研究方向为计算机信号处理、机器学习|张鹏 (1985—), 男, 本科, 主要研究方向为目标跟踪、图像处理以及计算机视觉|朱川 (1986—), 男, 本科, 主要研究方向为目标跟踪、图像检测与跟踪
基金资助:
国家自然科学基金(61773397);国家自然科学基金(61703423);国家自然科学基金(61701524);宁波自然科学基金(2021J049)

Temporal regularized correlation filter tracking algorithm based on multi-model distillation

Zhuling QIU¹, Yufei ZHA^2,^3,*, Zhenyu LI¹, Yuming LI¹, Peng ZHANG¹, Chuan ZHU¹

1. Unit 63787 of the PLA, Shihezi 832099, China
2. School of Computer Science, Northwestern Polytechnical University, Xi'an 710072, China
3. Ningbo Institute of Northwestern Polytechnical University, Ningbo 315000, China

Received:2021-08-24 Online:2022-08-01 Published:2022-08-24
Contact: Yufei ZHA

摘要/Abstract

摘要：

目前大多数基于相关滤波的跟踪方法是通过对模型采取简单的线性加权融合或是将历史模型作为时间正则化项来约束模型更新的方式, 增强滤波器对目标的判别能力, 但这种方式对目标时域信息利用有限, 容易造成模型退化漂移。本文提出一种基于多模型蒸馏的时间正则化相关滤波跟踪算法, 该方法通过收集跟踪过程中利用当前样本产生的独立模型, 在建立包含背景信息的局部样本库中来指导滤波器更新, 以此保留目标在时域中的鲁棒特征。同时，根据每一个模型对当前目标的表征能力不同进行可靠性权值更新。最后，利用交替方向乘子(alternating direction multiplier，ADMM)算法进行模型迭代优化。通过在大量的数据库进行实验, 结果表明本文的方法在精确度与成功率上有了大幅提升。

关键词: 目标跟踪, 相关滤波, 蒸馏学习, 时间正则化

Abstract:

At present, most correlation filter based tracking methods adopts simple linear weighted fusion of the model or use the historical model as the temporal regularization term to constrain the model update, which can enhance the ability of the filter to discriminate the target. However, this method cannot make full use of the information of the target, which is easy to cause model degradation and drift. This paper proposes a temporal regularized correlation filter based on multi-model distillation for visual tracking. This method collects the independent model generated by the current sample in the tracking process, which can guide the filter update in the local sample library containing background information. This can retain the robust features of the target in the temporal domain. At the same time, the reliability weight is updated according to the different characterization ability of each model for the current target. Finally, the alternating direction multiplier (ADMM) algorithm is used to iteratively optimize the model. A large number of experimental results in the databases show that the precision and success rate of the method have been greatly improved.

Key words: target tracking, correlation filter, distillation learning, temporal regularization

中图分类号:

TN919.82

仇祝令, 查宇飞, 李振宇, 李禹铭, 张鹏, 朱川. 基于多模型蒸馏的时间正则化相关滤波跟踪算法[J]. 系统工程与电子技术, 2022, 44(8): 2448-2456.

Zhuling QIU, Yufei ZHA, Zhenyu LI, Yuming LI, Peng ZHANG, Chuan ZHU. Temporal regularized correlation filter tracking algorithm based on multi-model distillation[J]. Systems Engineering and Electronics, 2022, 44(8): 2448-2456.

图/表 9

图1

图2

图3

图4

图5

表1

图6

图7

图8

参考文献 30

1	黄月平, 李小锋, 杨小冈, 等. 基于相关滤波的视觉目标跟踪算法新进展[J]. 系统工程与电子技术, 2021, 43 (8): 2051- 2065.
	HUANG Y P , LI X F , YANG X G , et al. Advances in visual object tracking algorithm based on correlation filter[J]. Systems Engineering and Electronics, 2021, 43 (8): 2051- 2065.
2	谢瑜, 陈莹. 空间注意机制下的自适应目标跟踪[J]. 系统工程与电子技术, 2019, 41 (9): 1945- 1954.
	XIE Y , CHEN Y . Adaptive object tracking based on spatial attention mechanism[J]. Systems Engineering and Electronics, 2019, 41 (9): 1945- 1954.
3	张红颖, 胡文博. 多特征融合的尺度自适应相关滤波跟踪算法[J]. 系统工程与电子技术, 2019, 41 (5): 951- 957.
	ZHANG H Y , HU W B . Scale-adaptive correlation filter tracking based on multiple features[J]. Systems Engineering and Electronics, 2019, 41 (5): 951- 957.
4	周治国, 荆朝, 王秋伶, 等. 基于时空信息融合的无人艇水面目标检测跟踪[J]. 电子与信息学报, 2021, 43 (6): 1698- 1705.
	ZHOU Z G , JING Z , WANG Q L , et al. Object detection and tracking of unmanned surface vehicles based on spatial-temporal information fusion[J]. Journal of Electronics and Information Technology, 2021, 43 (6): 1698- 1705.
5	陈丹, 姚伯羽. 运动模型引导的自适应核相关目标跟踪方法[J]. 电子学报, 2021, 49 (3): 550- 558.
	CHEN D , YAO B Y . Adaptive response kernel correlation target tracking method guided by motion model[J]. Acta Electronica Sinica, 2021, 49 (3): 550- 558.
6	刘宗达, 董立泉, 赵跃进, 等. 视频中快速运动目标的自适应模型跟踪算法[J]. 光学学报, 2021, 41 (18): 1815001.
	LIU Z D , DONG L Q , ZHAO Y J , et al. Adaptive model tracking algorithm for fast-moving targets in video[J]. Acta Optica Sinica, 2021, 41 (18): 1815001.
7	BOLME D S, BEVERIDGE J R, DRAPER B A, et al. Visual object tracking using adaptive correlation filters[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2010: 2544-2550.
8	HENRIQUES J F , CASEIRO R , MARTINS P , et al. High-speed tracking with kernelized correlation filters[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2014, 37 (3): 583- 596.
9	LI F, TIAN C, ZUO W M, et al. Learning spatial-temporal regularized correlation filters for visual tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4904-4913.
10	XU T Y , FENG Z H , WU X J , et al. Learning adaptive discriminative correlation filters via temporal consistency preserving spatial feature selection for robust visual object tracking[J]. IEEE Trans.on Image Processing, 2019, 28 (11): 5596- 5609. doi: 10.1109/TIP.2019.2919201
11	ZHU Z, WU W, ZOU W, et al. End-to-end flow correlation tracking with spatial-temporal attention[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 548-557.
12	DANELLJAN M, HAGER G, KHAN F S, et al. Learning spatially regularized correlation filters for visual tracking[C]//Proc. of the IEEE International Conference on Computer Vision, 2015: 4310-4318.
13	DANELLJAN M, HAGER G, SHAHBAZ K F, et al. Adaptive decontamination of the training set: a unified formulation for discriminative visual tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 1430-1438.
14	MUELLER M, SMITH N, GHANEM B. Context-aware correlation filter tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 1396-1404.
15	DAI K N, WANG D, LU H C, et al. Visual tracking via adaptive spatially-regularized correlation filters[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2019: 4670-4679.
16	KIANI G H, FAGG A, LUCEY S. Learning background-aware correlation filters for visual tracking[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 1763-1771.
17	DANELLJAN M , HÄGER G , KHAN F S , et al. Discriminative scale space tracking[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2016, 39 (8): 1561- 1575.
18	LI Y M, FU C H, DING F Q, et al. Autotrack: towards high-performance visual tracking for uav with automatic spatio-temporal regularization[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2020: 11923-11932.
19	MUELLER M, SMITH N, GHANEM B. A benchmark and simulator for uav tracking[C]//Proc. of the European Confe-rence on Computer Vision, 2016: 445-461.
20	DU D W, QI Y K, YU H Y, et al. The unmanned aerial vehicle benchmark: Object detection and tracking[C]//Proc. of the European Conference on Computer Vision, 2018: 370-386.
21	WU Y , LIM J , YANG M H . Object tracking benchmark[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2015, 37 (9): 1834- 1848. doi: 10.1109/TPAMI.2014.2388226
22	XU T, FENG Z H, WU X J, et al. Joint group feature selection and discriminative filter learning for robust visual object tracking[C]//Proc. of the IEEE International Conference on Computer Vision, 2019: 7950-7960.
23	LUKEZIC A, VOJIR T, CEHOVIN Z L, et al. Discriminative correlation filter with channel and spatial reliability[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 6309-6318.
24	BERTINETTO L, VALMADRE J, GOLODETZ S, et al. Staple: complementary learners for real-time tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 1401-1409.
25	NAM H, HAN B. Learning multi-domain convolutional neural networks for visual tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 4293-4302.
26	SONG Y B, MA C, GONG L J, et al. Crest: convolutional residual learning for visual tracking[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 2555-2564.
27	BERTINETTO L, VALMADRE J, HENRIQUES J F, et al. Fully-convolutional siamese networks for object tracking[C]//Proc. of the European Conference on Computer Vision, 2016: 850-865.
28	ZHANG J M, MA S G, SCLAROFF S. MEEM: robust tracking via multiple experts using entropy minimization[C]//Proc. of the European Conference on Computer Vision, 2014: 188-203.
29	KALAL Z , MIKOLAJCZYK K , MATAS J . Tracking-learning-detection[J]. IEEE Trans.on Pattern Analysis and Machine Intelligence, 2011, 34 (7): 1409- 1422.
30	DANELLJAN M, BHAT G, SHAHBAZ K F, et al. Efficient convolution operators for tracking[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 6638-6646.

Trackers	AUC/%	DP/%	Speed/FPS
baseline	45.9	66.0	35
baseline_sample	47.1	68.3	31
baseline_FWM	47.9	70.2	29
baseline_UWM	47.3	68.8	28
baseline_PUWM	48.7	72.5	25
Our	51.8	75.5	20

[1]	侯子林, 程婷, 彭瀚. 基于量测转换序贯滤波的GMPHD机动目标跟踪[J]. 系统工程与电子技术, 2022, 44(8): 2474-2482.
[2]	史浩然, 卢发兴, 祁江鑫, 杨光. 基于辅助信标的无人机协同目标跟踪[J]. 系统工程与电子技术, 2022, 44(7): 2302-2310.
[3]	金国栋, 薛远亮, 谭力宁, 许剑锟. 基于孪生神经网络的目标跟踪算法进展研究[J]. 系统工程与电子技术, 2022, 44(6): 1805-1822.
[4]	翟光, 王妍欣, 孙一勇. 基于低轨星网的多目标协同跟踪滤波技术[J]. 系统工程与电子技术, 2022, 44(6): 1957-1967.
[5]	王帅, 向建军, 彭芳, 唐书娟. 基于新最速下降法的目标跟踪算法[J]. 系统工程与电子技术, 2022, 44(5): 1512-1519.
[6]	辛怀声, 曹晨. 基于交互多模型的分组δ-广义标签多伯努利算法[J]. 系统工程与电子技术, 2022, 44(4): 1128-1138.
[7]	谢家豪, 黄树彩, 韦道知, 张曌宇, 王文豪. 基于P_EV准则的不确定混合多传感器联盟求解[J]. 系统工程与电子技术, 2022, 44(3): 819-826.
[8]	宋子壮, 杨嘉伟, 张东方, 王诗强, 张硕. 基于无锚框的红外多类别多目标实时跟踪网络[J]. 系统工程与电子技术, 2022, 44(2): 401-409.
[9]	姚云翔, 陈莹. 注意力机制下双模态交互融合的目标跟踪网络[J]. 系统工程与电子技术, 2022, 44(2): 410-419.
[10]	辛怀声, 宋鹏汉, 曹晨. 多模型广义标签多伯努利滤波器[J]. 系统工程与电子技术, 2022, 44(12): 3603-3613.
[11]	刘浩楠, 宋骊平. 基于核Fisher判别的群结构更新模型及群目标跟踪算法[J]. 系统工程与电子技术, 2022, 44(10): 3012-3019.
[12]	方澄, 路稳, 姬菁颖, 宋玉蒙, 梁斐菲, 罗志伟. 基于外观相似性更新的相关滤波跟踪算法[J]. 系统工程与电子技术, 2022, 44(1): 117-126.
[13]	黄月平, 李小锋, 杨小冈, 齐乃新, 卢瑞涛, 张胜修. 基于相关滤波的视觉目标跟踪算法新进展[J]. 系统工程与电子技术, 2021, 43(8): 2051-2065.
[14]	赵跃新, 齐望东, 刘鹏, 袁恩, 徐兵. 三维AoA目标跟踪的二次约束卡尔曼滤波算法[J]. 系统工程与电子技术, 2021, 43(8): 2263-2272.
[15]	孙明杰, 周林, 顾金玲, 李培庚. 基于多域网络的红外目标跟踪算法[J]. 系统工程与电子技术, 2021, 43(5): 1176-1183.

基于多模型蒸馏的时间正则化相关滤波跟踪算法

Temporal regularized correlation filter tracking algorithm based on multi-model distillation

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 30

相关文章 15

编辑推荐

Metrics

本文评价