Robust visual tracking through hedging multiple features
PeerJ Computer Science · 07.05.2026 · 12 · e3851
DOI: 10.7717/peerj-cs.3851 ↗Kurz gefasst
Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 07.05.2026. Verfasst von Xianyou Zeng und 5 weiteren Autor:innen. Die Arbeit nennt 102 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of the Anhui Provincial Department of Education, Zhejiang Provincial Science and Technology Project.
Abstract (Originalsprache)
Visual tracking is a challenging task because target objects often undergo significant appearance changes due to factors such as heavy occlusion, background clutter, deformation, and sudden movements. In this study, we propose a novel tracker that enhances discrimination between the target and the background by hedging multiple features, including handcrafted features and deep features extracted from various convolutional layers. A weak tracker is generated by applying correlation filters to each feature, and all weak trackers are hedged to create a stronger, more robust tracker. To achieve robust visual tracking, we employ a hedging approach that adaptively determines the weights of weak trackers by considering both the discrepancies between their historical and instantaneous performance and the variations among all weak trackers over time. Additionally, an adaptive window is utilized to emphasize the target region and mitigate boundary effects. Extensive experiments conducted on various benchmark datasets demonstrate the effectiveness and feasibility of the proposed approach. The source code for this study is openly available and can be accessed at the following link: https://github.com/xianyou916/hmft .
Förderung
- Natural Science Foundation of the Anhui Provincial Department of Education
- Zhejiang Provincial Science and Technology Project
Zitieren
APA
Zeng, X., Xu, L., Wang, H., Xu, C., Xu, Y., & Chen, X. (2026). Robust visual tracking through hedging multiple features. PeerJ Computer Science, 12, e3851. https://doi.org/10.7717/peerj-cs.3851
Vancouver
Zeng X, Xu L, Wang H, Xu C, Xu Y, Chen X. Robust visual tracking through hedging multiple features. PeerJ Computer Science. 2026;12:e3851. doi:10.7717/peerj-cs.3851
BibTeX
@article{Zeng2026,
author = {Zeng, Xianyou and Xu, Long and Wang, Hengyou and Xu, Chengzhen and Xu, Yonghong and Chen, Xingyu},
title = {Robust visual tracking through hedging multiple features},
journal = {PeerJ Computer Science},
year = {2026},
volume = {12},
pages = {e3851},
doi = {10.7717/peerj-cs.3851}
}
EUVIMED