密度峰值聚类的全尺度相关滤波跟踪方法

doi:10.3778/j.issn.1002-8331.2109-0333

摘要/Abstract

摘要： 针对背景感知相关滤波算法对目标进行尺度预测时采用图像金字塔，在跟踪过程中需要根据多尺度的目标样本训练出跟踪滤波器，导致跟踪过程中存在计算开销大、跟踪速度慢的问题，提出了一种基于密度峰值聚类的全尺度跟踪方法。分别提取搜索区域中目标和背景的特征信息，通过密度峰值聚类方法分别对目标和背景进行聚类；通过单尺度的滤波器对目标位置进行粗预测，并通过交替方向乘子法（ADMM）降低滤波器训练的时间复杂度；对搜索区域中的前景点和背景点进行分类，并根据尺度置信度得到目标的最终位置与尺度。该算法和目前一些主流的跟踪算法在公共数据集通过在OTB2013、OTB2015和DTB70上进行实验，在有效提高跟踪速度的前提下，面对旋转、遮挡等多种复杂情况时的跟踪效果较好、跟踪成功率较高，满足实时性要求。

关键词: 目标跟踪, 相关滤波, 密度峰值聚类, 计算机视觉

Abstract: Since image pyramid is adopted in scale prediction of targets with background aware correlation filtering algorithm, it’s required to train a tracking filter with multi-scale target samples during tracking, resulting in excessive computation and slow tracking speed. In order to solve these problems, a full-scale tracking method is proposed based on density peak clustering. First, the feature information of target and background in the search area is extracted, followed by clustering of target and background using density peak clustering; second, the target location is roughly predicted with single-scale filter, after which alternating direction method of multipliers（ADMM） is used to reduce the time complexity of filter training; finally, the foreground points and background points in the search area are categorized to obtain final location and scale of target based on scale confidence. According to the experiments conducted for the proposed algorithm and current mainstream tracking algorithm on OTB2013, OTB2015 and DTB70 with public data sets, the proposed algorithm achieves better tracking effect and higher success rate in face of complex circumstances such as rotation and blocking in addition to effective improvement of tracking speed, thus satisfying the requirement on real-time performance.

Key words: target tracking, correlation filtering, density peak clustering, computer vision

肖振久, 李鑫. 密度峰值聚类的全尺度相关滤波跟踪方法[J]. 计算机工程与应用, 2023, 59(5): 131-139.

XIAO Zhenjiu, LI Xin. Full-Scale Correlation Filtering Tracking Method Based on Density Peak Clustering[J]. Computer Engineering and Applications, 2023, 59(5): 131-139.

参考文献

[1] 葛宝义，左宪章，胡永江.视觉目标跟踪方法研究综述[J].中国图象图形学报，2018，23（8）：1091-1107.
GE Baoyi，ZUO Xianzhang，HU Yongjiang.A review of research on visual target tracking methods[J].Journal of Image and Graphics，2018，23（8）：1091-1107.
[2] FELDHEGE F，RICHTER K，BRUHN S，et al.MATLAB-based tools for automated processing of motion tracking data provided by the GRAIL[J].Gait & Posture，2021，90：422-426.
[3] CHRYSOS G G，ANTONAKOS E，SNAPE P，et al.A comprehensive performance evaluation of deformable face tracking “in-the-wild”[J].International Journal of Computer Vision，2018，126（2）：198-232.
[4] PREENA R R E，KALAIVANI D.Kernelized correlation filters used for high-speed tracking[J].Journal of Critical Reviews，2020，7：1461-1470.
[5] 朱文青，刘艳，卞乐，等.基于生成式模型的目标跟踪方法综述[J].微处理机，2017，38（1）：41-47.
ZHU Wenqing，LIU Yan，BIAN Le，et al.Overview of target tracking methods based on generative models[J].Microprocessor，2017，38（1）：41-47.
[6] KUMAR P M A，KAVITHA D，KUMAR S A.A hybrid generative-discriminative model for abnormal event detection in surveillance video scenes[J].International Journal of Information and Computer Security，2020，12（2/3）：253-268.
[7] MA X.Fixed-point tracking of English reading text based on mean shift and multi-feature fusion[J].Journal of Intelligent & Fuzzy Systems，2021，40（4）：6315-6325.
[8] HARE S，GOLODETZ S，SAFFARI A，et al.Struck：structured output tracking with kernels[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，38（10）：2096-2109.
[9] KALAL Z，MIKOLAJCZYK K，MATAS J.Tracking-learning-detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2011，34（7）：1409-1422.
[10] BOLME D S，BEVERIDGE J R，DRAPER B A，et al.Visual object tracking using adaptive correlation filters[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition，2010：2544-2550.
[11] HENRIQUES J F，CASEIRO R，MARTINS P，et al.Exploiting the circulant structure of tracking-by-detection with kernels[C]//European Conference on Computer Vision.Berlin，Heidelberg：Springer，2012：702-715.
[12] HENRIQUES J F，CASEIRO R，MARTINS P，et al.High-speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2014，37（3）：583-596.
[13] DALAL N，TRIGGS B.Histograms of oriented gradients for human detection[C]//2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition（CVPR’05），2005：886-893.
[14] KIANI GALOOGAHI H，FAGG A，LUCEY S.Learning background-aware correlation filters for visual tracking[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：1135-1143.
[15] 杨佳霖，王文伟，熊晓璇，等.快速尺度估计的增强型多核相关滤波算法[J].计算机工程与应用，2020，56（7）：210-220.
YANG Jialin，WANG Wenwei，XIONG Xiaoxuan，et al.Fast scale estimation based on enhanced multi-kernel correlation filtering algorithm[J].Computer Engineering and Applications，2020，56（7）：210-220.
[16] 任俊丽，郭浩，董亚飞，等.自适应尺度突变目标跟踪[J].中国图象图形学报，2020，25（6）：1150-1159.
REN Junli，GUO Hao，DONG Yafei，et al.Adaptive scale sudden change object tracking[J].Journal of Image and Graphics，2020，25（6）：1150-1159.
[17] 茅正冲，陈海东.自适应尺度的上下文感知相关滤波跟踪算法[J].计算机工程与应用，2021，57（3）：168-174.
MAO Zhengchong，CHEN Haidong.Adaptive scale context-aware correlation filter tracking algorithm[J].Computer Engineering and Applications，2021，57（3）：168-174.
[18] RODRIGUEZ A，LAIO A.Clustering by fast search and find of density peaks[J].Science，2014，344（6191）：1492-1496.
[19] SHERMAN J，MORRISON W J.Adjustment of an inverse matrix corresponding to a change in one element of a given matrix[J].The Annals of Mathematical Statistics，1950，21（1）：124-127.
[20] DANELLJAN M，HAGER G，SHAHBAZ KHAN F，et al.Learning spatially regularized correlation filters for visual tracking[C]//Proceedings of the IEEE International Conference on Computer Vision，2015：4310-4318.
[21] BERTINETTO L，VALMADRE J，GOLODETZ S，et al.Staple：complementary learners for real-time tracking[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：1401-1409.
[22] DAI K，WANG D，LU H，et al.Visual tracking via adaptive spatially-regularized correlation filters[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2019：4670-4679.
[23] LI Y，ZHU J，HOI S C H，et al.Robust estimation of similarity transformation for visual object tracking[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2019：8666-8673.
[24] LI Y，FU C，DING F，et al.Autotrack：towards high-performance visual tracking for uav with automatic spatio-temporal regularization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2020：11923-11932.
[25] DANELLJAN M，H?GER G，KHAN F S，et al.Discriminative scale space tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2016，39（8）：1561-1575.