Title
High speed long-term visual object tracking algorithm for real robot systems
Abstract
•Our tracker judges the confidence and it takes relocation module to recover when tracking fails.•Position drift is constantly suppressed by an appearance filter designed to keep the memory of target. That promotes the accuracy.•The lost target can be relocated wherever it reappears and whatever it changes by multi-scale sliding window.•Improving the performance and saving the resources. our tracker is practical for long-term real-time actual robot systems in which resources are saved.
Year
DOI
Venue
2021
10.1016/j.neucom.2020.12.113
Neurocomputing
Keywords
DocType
Volume
Long-term UAV tracking,Correlation filter,Drift correction,Target relocate
Journal
434
ISSN
Citations 
PageRank 
0925-2312
0
0.34
References 
Authors
0
5
Name
Order
Citations
PageRank
Muxi Jiang100.34
Rui Li2162.45
Qisheng Liu300.34
Yingjing Shi4101.91
Esteban Tlelo-Cuautle500.34