Low-Rank Multi-Channel Features for Robust Visual Object Tracking

被引:8
|
作者
Fawad [1 ]
Khan, Muhammad Jamil [1 ]
Rahman, MuhibUr [2 ]
Amin, Yasar [1 ]
Tenhunen, Hannu [3 ,4 ]
机构
[1] Univ Engn & Technol Taxila, ACTSENA Res Grp, Telecommun Engn Dept, Punjab 47050, Pakistan
[2] Polytech Montreal, Dept Elect Engn, Montreal, PQ H3T 1J4, Canada
[3] Royal Inst Technol KTH, Dept Elect Syst, Isafjordsgatan 26, SE-16440 Stockholm, Sweden
[4] Univ Turku, Dept Informat Technol, TUCS, FIN-20520 Turku, Finland
来源
SYMMETRY-BASEL | 2019年 / 11卷 / 09期
关键词
circulant matrix; texture; tracking; convolution; filter; RECOGNITION;
D O I
10.3390/sym11091155
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
Kernel correlation filters (KCF) demonstrate significant potential in visual object tracking by employing robust descriptors. Proper selection of color and texture features can provide robustness against appearance variations. However, the use of multiple descriptors would lead to a considerable feature dimension. In this paper, we propose a novel low-rank descriptor, that provides better precision and success rate in comparison to state-of-the-art trackers. We accomplished this by concatenating the magnitude component of the Overlapped Multi-oriented Tri-scale Local Binary Pattern (OMTLBP), Robustness-Driven Hybrid Descriptor (RDHD), Histogram of Oriented Gradients (HoG), and Color Naming (CN) features. We reduced the rank of our proposed multi-channel feature to diminish the computational complexity. We formulated the Support Vector Machine (SVM) model by utilizing the circulant matrix of our proposed feature vector in the kernel correlation filter. The use of discrete Fourier transform in the iterative learning of SVM reduced the computational complexity of our proposed visual tracking algorithm. Extensive experimental results on Visual Tracker Benchmark dataset show better accuracy in comparison to other state-of-the-art trackers.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] Robust visual tracking via incremental low-rank features learning
    Zhang, Changcheng
    Liu, Risheng
    Qiu, Tianshuang
    Su, Zhixun
    NEUROCOMPUTING, 2014, 131 : 237 - 247
  • [2] Robust Visual Tracking Based on Multi-channel Compressive Features
    Xu, Jianqiang
    Lu, Yao
    MULTIMEDIA MODELING (MMM 2017), PT I, 2017, 10132 : 341 - 352
  • [3] Low-Rank Sparse Learning for Robust Visual Tracking
    Zhang, Tianzhu
    Ghanem, Bernard
    Liu, Si
    Ahuja, Narendra
    COMPUTER VISION - ECCV 2012, PT VI, 2012, 7577 : 470 - 484
  • [4] Robust low-rank decomposition of multi-channel feature matrices for fabric defect detection
    Chunlei Li
    Chaodie Liu
    Guangshuai Gao
    Zhoufeng Liu
    Yuping Wang
    Multimedia Tools and Applications, 2019, 78 : 7321 - 7339
  • [5] Robust low-rank decomposition of multi-channel feature matrices for fabric defect detection
    Li, Chunlei
    Liu, Chaodie
    Gao, Guangshuai
    Liu, Zhoufeng
    Wang, Yuping
    MULTIMEDIA TOOLS AND APPLICATIONS, 2019, 78 (06) : 7321 - 7339
  • [6] Low-Rank Representation with Graph Constraints for Robust Visual Tracking
    Liu, Jieyan
    Ma, Ao
    Li, Jingjing
    Lu, Ke
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2017, E100D (06) : 1325 - 1338
  • [7] Object tracking via dense SIFT features and low-rank representation
    Yong Wang
    Xinbin Luo
    Lu Ding
    Jingjing Wu
    Soft Computing, 2019, 23 : 10173 - 10186
  • [8] Object tracking via dense SIFT features and low-rank representation
    Wang, Yong
    Luo, Xinbin
    Ding, Lu
    Wu, Jingjing
    SOFT COMPUTING, 2019, 23 (20) : 10173 - 10186
  • [9] Robust Structural Low-Rank Tracking
    Javed, Sajid
    Mahmood, Arif
    Dias, Jorge
    Werghi, Naoufel
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2020, 29 : 4390 - 4405
  • [10] Robust Visual Tracking Via Online Low-rank Sparse Representation
    Kong F.
    Wang D.
    Shen Q.
    Bian C.
    Yan X.
    Gongcheng Kexue Yu Jishu/Advanced Engineering Sciences, 2017, 49 (04): : 151 - 157