Asymmetric discriminative correlation filters for visual tracking

LI Shui-wang; JIANG Qian-bo; ZHAO Qi-jun; LU Li; FENG Zi-liang

doi:10.1631/FITEE.1900507

Your Location：

Home >

Browse articles >

Asymmetric discriminative correlation filters for visual tracking

Regular Papers | Updated：2022-05-19

- Asymmetric discriminative correlation filters for visual tracking
- 用于视频跟踪的非对称判别相关滤波器
- Frontiers of Information Technology & Electronic Engineering Vol. 21, Issue 10, Pages: 1467-1484(2020)
- Affiliations：
  
  National Key Laboratory of Fundamental Science on Synthetic Vision, College of Computer Science, Sichuan University, Chengdu 610065, China
- Author bio：
  
  [ "Shui-wang LI, E-mail:lishuiwang0721@163.com" ]
  [ "Qian-boJIANG, E-mail:jqianbo@163.com" ]
  [ "Qi-jun ZHAO, E-mail:qjzhao@scu.edu.cn" ]
  Li LU, E-mail:luli@scu.edu.cn
  [ "Zi-liang FENG, E-mail:fengziliang@scu.edu.cn" ]
- Funds：
- DOI：10.1631/FITEE.1900507
  CLC： TP391
- Received：20 September 2019，
  
  Revised：2020-;9-;2，
  
  Published：2020-10
- Accepted：
Scan QR Code
LI Shui-wang, JIANG Qian-bo, ZHAO Qi-jun, et al. Asymmetric discriminative correlation filters for visual tracking[J]. Frontiers of Information Technology & Electronic Engineering, 2020, 21(10): 1467-1484.
DOI：

LI Shui-wang, JIANG Qian-bo, ZHAO Qi-jun, et al. Asymmetric discriminative correlation filters for visual tracking[J]. Frontiers of Information Technology & Electronic Engineering, 2020, 21(10): 1467-1484. DOI： 10.1631/FITEE.1900507.

摘要

判别相关滤波器（DCF）是视频跟踪领域一种有效方法，显著推动了视频跟踪领域进展。然而，卷积算子的对称性会带来计算上的问题，并破坏广义的平移等变性。针对前一问题，人们提出许多解决方法，但对后一问题不够重视。本文分析循环卷积的对称性带来的问题，提出一种非对称卷积运算，且证明这种运算具有弱的广义平移等变性。利用提出的卷积运算，构造一个非对称判别相关滤波跟踪器（ADCF）。它对目标的平移更加敏感，且其非对称性允许滤波器和输入样本有不同空域大小，这使得ADCF的计算复杂性，从滤波器参数数量不随输入样本增大而增加的意义上说，更加可控。且ADCF对应的正规矩阵具有两级块Toeplitz矩阵结构，利用该结构可设计时间复杂度为

(

log

)、空间复杂度为

(

)的矩阵-向量乘法。此外，有别于基于DCF的跟踪器，ADCF引进空域和时域正则化项，本质上不会增加计算复杂度。在4个公开基准数据集（OTB-2013，OTB-2015，VOT-2016和Temple-Color）和一个合成数据集上进行对比实验，结果表明所提方法取得最优视频跟踪性能。

Abstract

Discriminative correlation filters (DCF) are efficient in visual tracking and have advanced the field significantly. However

the symmetry of correlation (or convolution) operator results in computational problems and does harm to the generalized translation equivariance. The former problem has been approached in many ways

whereas the latter one has not been well recognized. In this paper

we analyze the problems with the symmetry of circular convolution and propose an asymmetric one

which as a generalization of the former has a weak generalized translation equivariance property. With this operator

we propose a tracker called the asymmetric discriminative correlation filter (ADCF)

which is more sensitive to translations of targets. Its asymmetry allows the filter and the samples to have different sizes. This flexibility makes the computational complexity of ADCF more controllable in the sense that the number of filter parameters will not grow with the sample size. Moreover

the normal matrix of ADCF is a block matrix with each block being a two-level block Toeplitz matrix. With this well-structured normal matrix

we design an algorithm for multiplying an

$$ N \times N $$

two-level block Toeplitz matrix by a vector with time complexity

$$ O(N{\rm log}N) $$

and space complexity

$$ O(N) $$

instead of

$$ O(N^2) $$

. Unlike DCF-based trackers

introducing spatial or temporal regularization does not increase the essential computational complexity of ADCF. Comparative experiments are performed on a synthetic dataset and four benchmarks

including OTB-2013

OTB-2015

VOT-2016

and Temple-Color

and the results show that our method achieves state-of-the-art visual tracking performance.

关键词

Keywords

references

L Bertinetto , , , J Valmadre , , , S Golodetz , , , 等 . . Staple: complementary learners for real-time tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2016 . . p.1401 - - 1409 . . DOI: 10.1109/CVPR.2016.156 http://doi.org/10.1109/CVPR.2016.156 . .

DS Bolme , , , JR Beveridge , , , BA Draper , , , 等 . . Visual object tracking using adaptive correlation filters . . Proc IEEE Computer Society Conf on Computer Vision and Pattern Recognition , , 2010 . . p.2544 - - 2550 . . DOI: 10.1109/CVPR.2010.5539960 http://doi.org/10.1109/CVPR.2010.5539960 . .

BY Chen , , , D Wang , , , PX Li , , , 等 . . Real-time 'actor-critic' tracking . . Proc 15 th European Conf on Computer Vision , , 2018 . . p.318 - - 334 . . DOI: 10.1007/978-3-030-01234-2\_20 http://doi.org/10.1007/978-3-030-01234-2\_20 . .

J Choi , , , HJ Chang , , , S Yun , , , 等 . . Attentional correlation filter network for adaptive visual tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2017 . . p.4828 - - 4837 . . DOI: 10.1109/CVPR.2017.513 http://doi.org/10.1109/CVPR.2017.513 . .

N Dalal , , , B Triggs . . Histograms of oriented gradients for human detection . . Proc IEEE Computer Society Conf on Computer Vision and Pattern Recognition , , 2005 . . p.886 - - 893 . . DOI: 10.1109/CVPR.2005.177 http://doi.org/10.1109/CVPR.2005.177 . .

M Danelljan , , , FS Khan , , , M Felsberg , , , 等 . . Adaptive color attributes for real-time visual tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2014 . . p.1090 - - 1097 . . DOI: 10.1109/CVPR.2014.143 http://doi.org/10.1109/CVPR.2014.143 . .

M Danelljan , , , G Hger , , , FS Khan , , , 等 . . Learning spatially regularized correlation filters for visual tracking . . Proc IEEE Int Conf on Computer Vision , , 2015 . . p.4310 - - 4318 . . DOI: 10.1109/ICCV.2015.490 http://doi.org/10.1109/ICCV.2015.490 . .

M Danelljan , , , G Hger , , , FS Khan , , , 等 . . Adaptive decontamination of the training set: a unified formulation for discriminative visual tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2016a . . p.1430 - - 1438 . . DOI: 10.1109/CVPR.2016.159 http://doi.org/10.1109/CVPR.2016.159 . .

M Danelljan , , , A Robinson , , , FS Khan , , , 等 . . Beyond correlation filters: learning continuous convolution operators for visual tracking . . Proc 14 th European Conf on Computer Vision , , 2016b . . p.472 - - 488 . . DOI: 10.1007/978-3-319-46454-1\_29 http://doi.org/10.1007/978-3-319-46454-1\_29 . .

M Danelljan , , , G Hger , , , FS Khan , , , 等 . . Discriminative scale space tracking . . IEEE Trans Patt Anal Mach Intell , , 2017a . . 39 ( ( 8 ): ): 1561 - - 1575 . . DOI: 10.1109/TPAMI.2016.2609928 http://doi.org/10.1109/TPAMI.2016.2609928 . .

M Danelljan , , , G Bhat , , , FS Khan , , , 等 . . ECO: efficient convolution operators for tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2017b . . p.6638 - - 6646 . . DOI: 10.1109/CVPR.2017.733 http://doi.org/10.1109/CVPR.2017.733 . .

XP Dong , , , JB Shen . . Triplet loss in Siamese network for object tracking . . Proc 15 th European Conf on Computer Vision , , 2018 . . p.459 - - 474 . . DOI: 10.1007/978-3-030-01261-8\_28 http://doi.org/10.1007/978-3-030-01261-8\_28 . .

HK Galoogahi , , , T Sim , , , S Lucey . . Multi-channel correlation filters . . Proc IEEE Int Conf on Computer Vision , , 2013 . . p.3072 - - 3079 . . DOI: 10.1109/ICCV.2013.381 http://doi.org/10.1109/ICCV.2013.381 . .

HK Galoogahi , , , A Fagg , , , S Lucey . . Learning background-aware correlation filters for visual tracking . . Proc IEEE Int Conf on Computer Vision , , 2017 . . p.1135 - - 1143 . . DOI: 10.1109/ICCV.2017.129 http://doi.org/10.1109/ICCV.2017.129 . .

JF Henriques , , , R Caseiro , , , P Martins , , , 等 . . High-speed tracking with kernelized correlation filters . . IEEE Trans Patt Anal Mach Intell , , 2015 . . 37 ( ( 3 ): ): 583 - - 596 . . DOI: 10.1109/TPAMI.2014.2345390 http://doi.org/10.1109/TPAMI.2014.2345390 . .

U Kart , , , A Lukezic , , , M Kristan , , , 等 . . Object tracking by reconstruction with view-specific discriminative correlation filters . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2019 . . p.1339 - - 1348 . . DOI: 10.1109/CVPR.2019.00143 http://doi.org/10.1109/CVPR.2019.00143 . .

M Kristan , , , A Leonardis , , , J Matas , , , 等 . . The visual object tracking VOT2016 challenge results . . Proc Amsterdam on Computer Vision , , 2016 . . p.191 - - 217 . . DOI: 10.1007/978-3-319-48881-3\_54 http://doi.org/10.1007/978-3-319-48881-3\_54 . .

D Lee . . Fast multiplication of a recursive block Toeplitz matrix by a vector and its application . . J Complex , , 1986 . . 2 ( ( 4 ): ): 295 - - 305 . . DOI: 10.1016/0885-064x(86)90007-5 http://doi.org/10.1016/0885-064x(86)90007-5 . .

B Li , , , JJ Yan , , , W Wu , , , 等 . . High performance visual tracking with Siamese region proposal network . . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , , 2018 . . p.8971 - - 8980 . . DOI: 10.1109/CVPR.2018.00935 http://doi.org/10.1109/CVPR.2018.00935 . .

F Li , , , C Tian , , , WM Zuo , , , 等 . . Learning spatial-temporal regularized correlation filters for visual tracking . . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , , 2018 . . p.4904 - - 4913 . . DOI: 10.1109/CVPR.2018.00515 http://doi.org/10.1109/CVPR.2018.00515 . .

Y Li , , , JK Zhu . . A scale adaptive kernel correlation filter tracker with feature integration . . Proc European Conf on Computer Vision , , 2014 . . p.254 - - 265 . . DOI: 10.1007/978-3-319-16181-5\_18 http://doi.org/10.1007/978-3-319-16181-5\_18 . .

PP Liang , , , E Blasch , , , HB Ling . . Encoding color information for visual tracking: algorithms and benchmark . . IEEE Trans Image Process , , 2015 . . 24 ( ( 12 ): ): 5630 - - 5644 . . DOI: 10.1109/TIP.2015.2482905 http://doi.org/10.1109/TIP.2015.2482905 . .

A Lukezic , , , T Voj , , , LC Zajc , , , 等 . . Discriminative correlation filter with channel and spatial reliability . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2017 . . p.4847 - - 4856 . . DOI: 10.1109/CVPR.2017.515 http://doi.org/10.1109/CVPR.2017.515 . .

C Ma , , , JB Huang , , , XK Yang , , , 等 . . Hierarchical convolutional features for visual tracking . . Proc IEEE Int Conf on Computer Vision , , 2015 . . p.3074 - - 3082 . . DOI: 10.1109/ICCV.2015.352 http://doi.org/10.1109/ICCV.2015.352 . .

M Mueller , , , N Smith , , , B Ghanem . . Context-aware correlation filter tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2017 . . p.1387 - - 1395 . . DOI: 10.1109/CVPR.2017.152 http://doi.org/10.1109/CVPR.2017.152 . .

H Nam , , , B Han . . Learning multi-domain convolutional neural networks for visual tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2016 . . p.4293 - - 4302 . . DOI: 10.1109/CVPR.2016.465 http://doi.org/10.1109/CVPR.2016.465 . .

S Pu , , , Y Song , , , C Ma , , , 等 . . Deep attentive tracking via reciprocative learning . . Proc 32 nd Conf on Neural Information Processing Systems , , 2018 . . p.1931 - - 1941 . . http://arxiv.org/abs/1810.03851 http://arxiv.org/abs/1810.03851 , , . .

YK Qi , , , SP Zhang , , , L Qin , , , 等 . . Hedged deep tracking . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2016 . . p.4303 - - 4311 . . DOI: 10.1109/CVPR.2016.466 http://doi.org/10.1109/CVPR.2016.466 . .

C Sun , , , D Wang , , , HC Lu , , , 等 . . Correlation tracking via joint discrimination and reliability learning . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2018 . . p.489 - - 497 . . DOI: 10.1109/CVPR.2018.00058 http://doi.org/10.1109/CVPR.2018.00058 . .

YX Sun , , , C Sun , , , D Wang , , , 等 . . ROI pooled correlation filters for visual tracking . . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , , 2019 . . p.5783 - - 5791 . . DOI: 10.1109/CVPR.2019.00593 http://doi.org/10.1109/CVPR.2019.00593 . .

M Tang , , , B Yu , , , F Zhang , , , 等 . . High-speed tracking with multi-kernel correlation filters . . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , , 2018 . . p.4874 - - 4883 . . DOI: 10.1109/CVPR.2018.00512 http://doi.org/10.1109/CVPR.2018.00512 . .

Q Wang , , , L Zhang , , , L Bertinetto , , , 等 . . Fast online object tracking and segmentation: a unifying approach . . Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition , , 2019 . . p.1328 - - 1338 . . DOI: 10.1109/CVPR.2019.00142 http://doi.org/10.1109/CVPR.2019.00142 . .

Y Wu , , , J Lim , , , MH Yang . . Online object tracking: a benchmark . . Proc IEEE Conf on Computer Vision and Pattern Recognition , , 2013 . . p.2411 - - 2418 . . DOI: 10.1109/CVPR.2013.312 http://doi.org/10.1109/CVPR.2013.312 . .

Y Wu , , , J Lim , , , MH Yang . . Object tracking benchmark . . IEEE Trans Patt Anal Mach Intell , , 2015 . . 37 ( ( 9 ): ): 1834 - - 1848 . . DOI: 10.3389/FNINS.2016.00405 http://doi.org/10.3389/FNINS.2016.00405 . .

JM Zhang , , , SG Ma , , , S Sclaroff . . MEEM: robust tracking via multiple experts using entropy minimization . . Proc 13 th European Conf on Computer Vision , , 2014 . . p.188 - - 203 . . DOI: 10.1007/978-3-319-10599-4\_13 http://doi.org/10.1007/978-3-319-10599-4\_13 . .

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

Chat

Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
Tel：+86-571-88273162 Email：fitee@zju.edu.cn
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰