期刊论文

Zhang, JM

英文

Computers & Electrical Engineering

0045-7906

2024

114

109075

10.1016/j.compeleceng.2024.109075

This work was supported in part by the National Natural Science Foundation of China under Grant 61972056, the Open Fund of Key Laboratory of Safety Control of Bridge Engineering, Ministry of Education (Changsha University of Science and Technology) under Grant 21KB06, and the Postgraduate Scientific Research Innovation Fund of Changsha University of Science and Technology under Grant CXCLY2022115. CRediT authorship contribution statement Jianming Zhang: Conceptualization, Formal analysis, Writing – review & editing, Supervision, acquisition. Yufan He: Methodology, Formal analysis, Software, Writing – original draft. Wentao Chen: Data curation, Visualization. Li-Dan Kuang: Formal analysis, Validation. Bin Zheng: Project administration, Investigation.

本校为第一且通讯机构

The fusion of the template and search region features plays a significant role in deep learningbased trackers. In Siamese -based trackers, different cross -correlation operations are commonly used to fuse features, which cannot obtain global connections. On the other hand, transformerbased trackers use attention mechanism to fuse features, which cannot suppress the interference of distractors in the background. Furthermore, existing trackers use regression and classification heads with the same structure, which leads to lack a deeper understanding of these two different tasks. To address these...