Next Article in Journal
Network Management and Monitoring Solutions for Vehicular Networks: A Survey
Next Article in Special Issue
Audio-Based Aircraft Detection System for Safe RPAS BVLOS Operations
Previous Article in Journal
Analog Realization of a Fractional Recursive Variable-Type and Order Operator for a Particular Switching Strategy
Open AccessArticle

A Learning Frequency-Aware Feature Siamese Network for Real-Time Visual Tracking

School of Software Engineering, Beijing Jiaotong University, No. 3 Shangyuancun, Haidian District, Beijing 100044, China
Author to whom correspondence should be addressed.
Electronics 2020, 9(5), 854;
Received: 16 April 2020 / Revised: 15 May 2020 / Accepted: 19 May 2020 / Published: 21 May 2020
(This article belongs to the Special Issue Deep Learning Technologies for Machine Vision and Audition)
Visual object tracking by Siamese networks has achieved favorable performance in accuracy and speed. However, the features used in Siamese networks have spatially redundant information, which increases computation and limits the discriminative ability of Siamese networks. Addressing this issue, we present a novel frequency-aware feature (FAF) method for robust visual object tracking in complex scenes. Unlike previous works, which select features from different channels or layers, the proposed method factorizes the feature map into multi-frequency and reduces the low-frequency information that is spatially redundant. By reducing the low-frequency map’s resolution, the computation is saved and the receptive field of the layer is also increased to obtain more discriminative information. To further improve the performance of the FAF, we design an innovative data-independent augmentation for object tracking to improve the discriminative ability of tracker, which enhanced linear representation among training samples by convex combinations of the images and tags. Finally, a joint judgment strategy is proposed to adjust the bounding box result that combines intersection-over-union (IoU) and classification scores to improve tracking accuracy. Extensive experiments on 5 challenging benchmarks demonstrate that our FAF method performs favorably against SOTA tracking methods while running around 45 frames per second. View Full-Text
Keywords: deep learning; computer version; object tracking deep learning; computer version; object tracking
Show Figures

Figure 1

MDPI and ACS Style

Yang, Y.; Xing, W.; Zhang, S.; Yu, Q.; Guo, X.; Guo, M. A Learning Frequency-Aware Feature Siamese Network for Real-Time Visual Tracking. Electronics 2020, 9, 854.

AMA Style

Yang Y, Xing W, Zhang S, Yu Q, Guo X, Guo M. A Learning Frequency-Aware Feature Siamese Network for Real-Time Visual Tracking. Electronics. 2020; 9(5):854.

Chicago/Turabian Style

Yang, Yuxiang; Xing, Weiwei; Zhang, Shunli; Yu, Qi; Guo, Xiaoyu; Guo, Min. 2020. "A Learning Frequency-Aware Feature Siamese Network for Real-Time Visual Tracking" Electronics 9, no. 5: 854.

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

Search more from Scilit
Back to TopTop