Convolutional Neural Network with Structural Input for Visual Object Tracking

被引:6
作者
Fiaz, Mustansar [1 ]
Mahmood, Arif [2 ]
Jung, Soon Ki [1 ]
机构
[1] Kyungpook Natl Univ, Sch Comp Sci & Engn, Daegu, South Korea
[2] Informat Technol Univ, Dept Comp Sci, Lahore, Pakistan
来源
SAC '19: PROCEEDINGS OF THE 34TH ACM/SIGAPP SYMPOSIUM ON APPLIED COMPUTING | 2019年
关键词
Deep learning; convolutional neural network; visual tracking; machine learning;
D O I
10.1145/3297280.3297416
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Numerous deep learning approaches have been applied to visual object tracking owing to their capabilities to leverage huge training data for performance improvement. Most of these approaches have limitations with regard to learning target specific information rich features and therefore observe reduced accuracy in the presence of different challenges such as occlusion, scale variations, rotation and clutter. We proposed a deep neural network that takes input in the form of two stacked patches and regresses both the similarity and the dis-similarity scores in single evaluation. Image patches are concatenated depth-wise and fed to a six channel input of the network. The proposed network is generic and exploits the structural differences between the two input patches to obtain more accurate similarity and dissimilarity scores. Online learning is enforced via short-term and long-term updates to improve the tracking performance. Extensive experimental evaluations have been performed on OTB2015 and TempleColor128 benchmark datasets. Comparisons with state-of-the-art methods indicate that the proposed framework has achieved better tracking performance. The proposed tracking framework has obtained improved accuracy in different challenges including occlusion, background clutter, in-plane rotation and scale variations.
引用
收藏
页码:1345 / 1352
页数:8
相关论文
共 56 条
[1]  
Adam A., 2006, IEEE C COMP VIS PATT, P798, DOI [DOI 10.1109/CVPR.2006.256, 10.1109/CVPR.2006.256]
[2]  
Alan Luke A, 2017, CVPR
[3]  
[Anonymous], MULTIMED TOOLS APPL
[4]  
Avidan S., 2004, IEEE T P A M I, V26, P8
[5]   Robust Object Tracking with Online Multiple Instance Learning [J].
Babenko, Boris ;
Yang, Ming-Hsuan ;
Belongie, Serge .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2011, 33 (08) :1619-1632
[6]   Fully-Convolutional Siamese Networks for Object Tracking [J].
Bertinetto, Luca ;
Valmadre, Jack ;
Henriques, Joao F. ;
Vedaldi, Andrea ;
Torr, Philip H. S. .
COMPUTER VISION - ECCV 2016 WORKSHOPS, PT II, 2016, 9914 :850-865
[7]  
Bolme DS, 2010, PROC CVPR IEEE, P2544, DOI 10.1109/CVPR.2010.5539960
[8]  
Brown M, 2017, INVENTING AGENCY: ESSAYS ON THE LITERARY AND PHILOSOPHICAL PRODUCTION OF THE MODERN SUBJECT, P17
[9]   BIT: Biologically Inspired Tracker [J].
Cai, Bolun ;
Xu, Xiangmin ;
Xing, Xiaofen ;
Jia, Kui ;
Miao, Jie ;
Tao, Dacheng .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (03) :1327-1339
[10]   Histograms of oriented gradients for human detection [J].
Dalal, N ;
Triggs, B .
2005 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOL 1, PROCEEDINGS, 2005, :886-893