多通道核相关滤波的实时跟踪方法

doi:10.11772/j.issn.1001-9081.2015.12.3544

计算机应用 ›› 2015, Vol. 35 ›› Issue (12): 3544-3549.DOI: 10.11772/j.issn.1001-9081.2015.12.3544

多通道核相关滤波的实时跟踪方法

胡昭华^1,2, 邢卫国¹, 何军¹, 张秀再^1,2

1. 南京信息工程大学电子与信息工程学院, 南京 210044;
2. 江苏省大气环境与装备技术协同创新中心, 南京 210044

收稿日期:2015-05-18 修回日期:2015-08-01 出版日期:2015-12-10 发布日期:2015-12-10
通讯作者: 胡昭华(1981-),女,安徽马鞍山人,副教授,博士,主要研究方向:视觉跟踪、模式识别
作者简介:邢卫国(1990-),男,江苏滨海人,硕士研究生,主要研究方向:目标跟踪;何军(1978-),男,河南郑州人,讲师,硕士,主要研究方向:计算机视觉、高维数据分析、压缩感知;张秀再(1979-),男,江苏沛县人,副教授,博士,主要研究方向:信号与信息处理。
基金资助:
国家自然科学基金资助项目(61203273);江苏省自然科学基金资助项目(BK20141004);江苏省普通高校自然科学研究项目(11KJB510009,14KJB510019);江苏高校优势学科Ⅱ期建设工程项目。

Real-time object tracking method based on multi-channel kernel correlation filter

HU Zhaohua^1,2, XING Weiguo¹, HE Jun¹, ZHANG Xiuzai^1,2

1. School of Electronic and Information Engineering, Nanjing University of Information Science and Technology, Nanjing Jiangsu 210044, China;
2. Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing Jiangsu 210044, China

Received:2015-05-18 Revised:2015-08-01 Online:2015-12-10 Published:2015-12-10

摘要/Abstract

摘要： 现有跟踪算法大都需要构建复杂的外观模型、抽取大量训练样本来实现精确的目标跟踪,会产生庞大的计算量,不利于实时跟踪。鉴于此,提出了一种多通道核相关滤波的实时跟踪方法。首先,利用核化岭回归方法对视频帧的目标信息进行训练,学习得到滤波模板;接着,用滤波模板对待检测帧的可能区域进行相关性度量;最后,将相关度最高的位置作为跟踪结果,并通过对多通道的独立输入进行加权求和,解决多通道输入问题。与现有跟踪方法的大量对比实验表明,在不同的挑战因素下,该方法在保证跟踪精度的同时,跟踪速度也存在明显优势。该方法通过相关滤波的方式可避免抽取大量样本,并利用频域的点乘代替时域的相关运算,大大降低了计算复杂度,使跟踪速度完全满足实时场景的跟踪需求。

关键词: 视觉跟踪, 相关滤波, 核函数, 实时性

Abstract: The most existing algorithms have to build the complex model and draw a large number of training samples to achieve accurate object tracking,which will produce large amount of calculation. The proposed problem is not conducive to real-time tracking. In order to solve the problem, a real-time tracking method based on multi-channel kernel correlation filter was presented. Firstly, the target information of video frames were trained by using the nucleation ridge regression method to get the filter template. Secondly, the filter template was utilized to carry out the correlation measure for the possible area of the frame to be detected. Finally, the most relevant location was considered as the tracking result and the independent inputs of multiple channels were weighted and then added to solve the problem of multi-channel input. A large number of comparison experiments with the existing tracking methods show that, the proposed method guarantees the tracking accuracy and its tracking speed also has obvious advantages under different challenge factors. The proposed method avoids to extract a large number of samples by the correlation filter and use the dot product of frequency domain to replace the correlation operation of time-domain, which greatly reduces the computational complexity and makes the tracking speed completely meet the tracking demand of real-time scenario.

Key words: visual tracking, correlation filter, kernel function, real-time

中图分类号:

TP391.41

胡昭华, 邢卫国, 何军, 张秀再. 多通道核相关滤波的实时跟踪方法[J]. 计算机应用, 2015, 35(12): 3544-3549.

HU Zhaohua, XING Weiguo, HE Jun, ZHANG Xiuzai. Real-time object tracking method based on multi-channel kernel correlation filter[J]. Journal of Computer Applications, 2015, 35(12): 3544-3549.

参考文献

[1] DANELLJAN M, KHAN F S, FELSBERG M, et al. Adaptive color attributes for real-time visual tracking[C]//CVPR 2014:Proceedings of the 2014 Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2014:1090-1097.
[2] HENRIQUES J F, CASEIRO R, MARTINS P. High-speed tracking with kernelized correlation filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 37(3):583-596.
[3] ZHANG K, ZHANG L, LIU Q, et al. Fast visual tracking via dense spatio-temporal context learning[C]//ECCV 2014:Proceedings of the 13th European Conference on Computer Vision, LNCS 8693. Berlin:Springer, 2014:127-141.
[4] PENG S, PENG X. Object tracking with efficient multiple instance learning[J]. Journal of Computer Applications, 2015, 35(2):466-469. (彭爽,彭晓明.基于高效多示例学习的目标跟踪[J].计算机应用,2015,35(2):466-469.)
[5] ZHONG W. Robust object tracking via sparsity-based collaborative model[C]//CVPR 2012:Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2012:1838-1845.
[6] ROSS D, LIM J, LIN R-S, et al.Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2008, 77(1/2/3):125-141.
[7] MENG J, LIU J, HAN M. Marginalized particle filter for combined feature target-tracking[J]. Journal of Application Research of Computers, 2015, 32(7):135-144. (孟军英,刘教民,韩明.基于联合特征的边缘粒子滤波目标跟踪算法研究[J].计算机应用研究,2015,32(7):135-144.)
[8] YANG F, LU H, YANG M. Robust super-pixel tracking[J]. IEEE Transaction on Image Processing, 2014, 23(4):1639-1651.
[9] BABENKO B, YANG M-H, BELONGIE S. Robust object tracking with online multiple instance learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8):1619-1632.
[10] YANG F, LU H, YANG M. Robust visual tracking via multiple kernel boosting with affinity constraints[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2014, 24(2):242-254.
[11] ZHANG K, ZHANG L, YANG M. Real-time object tracking via online discriminative feature selection[J]. IEEE Transactions on Image Processing, 2013, 22(12):4664-4677.
[12] BOLME D, BEVERIDGE J, B DRAPER, et al.Visual object tracking using adaptive correlation filters[C]//CVPR 2010:Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2010:2544-2550.
[13] HENRIQUES J, CASEIRO R, MARTINS P, et al.Exploiting the circulant structure of tracking-by-detection with kernels[C]//ECCV 2012:Proceedings of the 2012 European Conference on Computer Vision. Berlin:Springer, 2012:702-715.
[14] RIFKIN R, YEO G, POGGIO T. Regularized least-squares classification[M]//Advances in Learning Theory:Methods, Models and Applications. Amsterdam:IOS Press, 2003:131-154.
[15] MESSERSCHMITT D G. Stationary points of a real-valued function of a complex variable, UCB/EECS-2006-93[R]. Berkeley:University of California, EECS Department, 2006:1-8
[16] KALAL Z, MIKOLAJCZYK K, MATAS J. Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7):1409-1422.
[17] JIA X, LU H, YANG M. Visual tracking via adaptive structural local sparse appearance model[C]//CVPR 2012:Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2012:1822-1829.
[18] ADAM A, RIVLIN E, SHIMSHONI I. Robust fragments-based tracking using the integral histogram[C]//CVPR 2006:Proceedings of the 2006 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2006:798-805.
[19] HARE S, SAFFARI A, TORR P H S. Struck:structured output tracking with kernels[C]//ICCV 2011:Proceedings of the 2011 International Conference on Computer Vision. Piscataway:IEEE, 2011:263-270.
[20] WU Y, LIM J, YANG M. Online object tracking:a benchmark[C]//CVPR 2013:Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2013:2411-2418.

多通道核相关滤波的实时跟踪方法

Real-time object tracking method based on multi-channel kernel correlation filter

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	刘晓光, 靳少康, 韦子辉, 梁铁, 王洪瑞, 刘秀玲. 基于阈值和极端随机树的实时跌倒检测方法[J]. 计算机应用, 2021, 41(9): 2761-2766.
[2]	姬张建, 任兴旺. 带旋转与尺度估计的全卷积孪生网络目标跟踪算法[J]. 计算机应用, 2021, 41(9): 2705-2711.
[3]	王宁, 宋慧慧, 张开华. 基于距离加权重叠度估计与椭圆拟合优化的精确目标跟踪算法[J]. 计算机应用, 2021, 41(4): 1100-1105.
[4]	効琦, 尹增山, 高爽. 基于检测与跟踪相互迭代的极暗弱目标搜索算法[J]. 计算机应用, 2021, 41(10): 3017-3024.
[5]	熊昌镇, 李言. 动态的加权孪生网络跟踪算法[J]. 计算机应用, 2020, 40(8): 2214-2218.
[6]	张杰, 常天庆, 戴文君, 郭理彬, 张雷. 基于相关滤波与颜色概率模型的目标跟踪算法[J]. 计算机应用, 2020, 40(6): 1774-1782.
[7]	徐小超, 严华. 引入目标分块模型的核相关滤波目标追踪算法[J]. 计算机应用, 2020, 40(3): 683-688.
[8]	费大胜, 宋慧慧, 张开华. 基于多层特征增强的实时视觉跟踪[J]. 计算机应用, 2020, 40(11): 3300-3305.
[9]	李旻择, 李小霞, 王学渊, 孙维. 基于多尺度核特征卷积神经网络的实时人脸表情识别[J]. 计算机应用, 2019, 39(9): 2568-2574.
[10]	贾永超, 何小卫, 郑忠龙. 融合重检测机制的卷积回归网络目标跟踪算法[J]. 计算机应用, 2019, 39(8): 2247-2251.
[11]	杨康, 宋慧慧, 张开华. 基于双重注意力孪生网络的实时视觉跟踪[J]. 计算机应用, 2019, 39(6): 1652-1656.
[12]	杨春德, 刘京, 瞿中. 分离窗口快速尺度自适应目标跟踪算法[J]. 计算机应用, 2019, 39(4): 1145-1149.
[13]	胡秀华, 王长元, 肖锋, 王亚文. 利用空间结构信息的相关滤波目标跟踪算法[J]. 计算机应用, 2019, 39(4): 1150-1156.
[14]	熊晓璇, 王文伟. 基于快速判别式尺度估计的核相关滤波方法[J]. 计算机应用, 2019, 39(2): 546-550.
[15]	闫若怡, 熊丹, 于清华, 肖军浩, 卢惠民. 基于并行跟踪检测框架与深度学习的目标跟踪算法[J]. 计算机应用, 2019, 39(2): 343-347.