Dual-template adaptive correlation filter for real-time object tracking
- PDF / 3,386,912 Bytes
- 22 Pages / 439.37 x 666.142 pts Page_size
- 28 Downloads / 218 Views
Dual-template adaptive correlation filter for real-time object tracking Junrong Yan 1 & Luchao Zhong 1 & Yingbiao Yao 1
1
& Xin Xu & Chenjie Du
1
Received: 6 November 2019 / Revised: 4 July 2020 / Accepted: 18 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Visual object tracking is a hot topic in the field of computer vision. The drift of the tracking box or loss of the tracking target often occurs for existing correlation filter based trackers when the target moves quickly or deforms. Focusing on this problem, we propose a dual-template adaptive correlation filter for real-time object tracking. First, we trained templates for different size levels. Second, the best template was selected based on the target response confidence during estimation of the target translation. Third, the dual templates, scale estimation component, and feature fusion component were integrated into the benchmark tracker, the kernelized correlation filter. The object tracking benchmark was used to evaluate the performance of the proposed algorithm. The experimental results show that compared with the benchmark tracker, the average overlap precision and distance precision of this proposed algorithm are increased by 23.2% and 9.4% in OTB-100. The average running frame rate reaches 42 frames per second, which can meet the real-time requirements. At the same time, five algorithms, DSST, SAMF, KCF, CN, and CSK, appear to drift or even lose the target among the four selected typical video sequences, while our algorithm can successfully track the target. Keywords Adaptive threshold . Correlation filter . Dual templates . Visual tracking
* Yingbiao Yao [email protected] Junrong Yan [email protected] Luchao Zhong [email protected] Xin Xu [email protected] Chenjie Du [email protected]
1
School of Communication Engineering, Hangzhou Dianzi University, Hangzhou 310018, China
Multimedia Tools and Applications
1 Introduction Visual object tracking is a popular topic in computer vision, with a wide range of applications in industrial fields (i.e., robots [32], surveillance systems). Although it has been studied for decades, it still has many challenges to solve. For example, the algorithm lacks prior information, and the tracker only uses the target information of the first frame. At the same time, there are many challenges in the video that affect the performance of the tracker, such as scale variations, illumination variations, etc. Therefore, it is challenging to design a robust tracking algorithm. Nowadays, visual object tracking algorithms can be divided into traditional tracking algorithms [4, 16, 17] and deep learning based tracking algorithms [1, 6, 11, 13, 18, 22, 28, 30, 38]. In general, trackers based on deep learning have better performance than traditional trackers. However, the former requires a large number of training samples. In addition, with the increase of the number of network layers, the computational complexity and parameter storage space are increased exponentially. Usually, tracker
Data Loading...
