EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Adaptive Multilevel Feature Fusion and Dynamic Model Updating for Robust Visual Object Tracking

Xianyou Zeng, Long Xu, Hengyou Wang, Lei Liu, Chengzhen Xu, Yonghong Xu, Xingyu Chen

Applied Computational Intelligence and Soft Computing · 2026-01 · 2026 (1)

DOI: 10.1155/acis/7770863 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Computational Intelligence and Soft Computing, online erschienen am 05.06.2026. Verfasst von Xianyou Zeng und 6 weiteren Autor:innen. Die Arbeit nennt 81 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Anhui Provincial Department of Education, Medical Science and Technology Project of Zhejiang Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Object tracking remains a challenging task due to significant variations in the appearance of target objects caused by factors such as deformation, occlusion, and rapid movement. These variations often result in model drift and tracking failure, particularly in complex and dynamic environments. To address these challenges, this study proposes an adaptive online tracking algorithm that fuses hierarchical convolutional features through adaptive weighting and employs a dynamic model‐updating mechanism. In the proposed approach, correlation filters are independently trained on each convolutional layer to generate multilevel response maps. These maps are then adaptively fused using learned weights to produce a final response map, from which the target position is determined based on the peak response. Additionally, a dynamic update strategy is designed to mitigate the degradation of the appearance model during long‐term tracking. Extensive experiments conducted on several public benchmark datasets demonstrate that the proposed tracker achieves superior robustness and accuracy compared to many state‐of‐the‐art methods, confirming its effectiveness for practical visual tracking applications.

Förderung

  • Anhui Provincial Department of Education
  • Medical Science and Technology Project of Zhejiang Province

Zitieren

APA

Zeng, X., Xu, L., Wang, H., Liu, L., Xu, C., Xu, Y., & Chen, X. (2026). Adaptive Multilevel Feature Fusion and Dynamic Model Updating for Robust Visual Object Tracking. Applied Computational Intelligence and Soft Computing, 2026(1). https://doi.org/10.1155/acis/7770863

Vancouver

Zeng X, Xu L, Wang H, Liu L, Xu C, Xu Y, et al. Adaptive Multilevel Feature Fusion and Dynamic Model Updating for Robust Visual Object Tracking. Applied Computational Intelligence and Soft Computing. 2026;2026(1). doi:10.1155/acis/7770863

BibTeX
@article{Zeng2026,
  author = {Zeng, Xianyou and Xu, Long and Wang, Hengyou and Liu, Lei and Xu, Chengzhen and Xu, Yonghong and Chen, Xingyu},
  title = {Adaptive Multilevel Feature Fusion and Dynamic Model Updating for Robust Visual Object Tracking},
  journal = {Applied Computational Intelligence and Soft Computing},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/acis/7770863}
}
Kostenlos starten →