Robust Tracking via Learning Model Update With Unsupervised Anomaly Detection Philosophy
IEEE Transactions on Circuits and Systems for Video Technology · 2023-05 · 33 (5) · 2330-2341
DOI: 10.1109/tcsvt.2022.3221723 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2023-05. Verfasst von Jie Gao und 2 weiteren Autor:innen. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 63 Literaturangaben. Gefördert von: Guangxi Science and Technology, National Natural Science Foundation of China, Nature Science Foundation of Fujian Province.
Abstract (Originalsprache)
Template tracking is a typical paradigm to adaptively locate arbitrary objects in the tracking literature. Although existing works present diverse template updating approaches, one of the essential problems of template updating has not been solved effectively, i.e., when and how to update a template. In this work, we treat the updating time as an abnormal moment that indicates the previous template cannot depict the target accurately any more. Thus, we introduce an effective State-Edge Awareness (SEA) module that detect such abnormal moments via unsupervised anomaly detection. To be specific, by retaining multi search frames of a video, SEA firstly analysis the correlation features that generated by the template and search images. Then, it estimates the measurement for abnormal degree that is regarded as the sign for template updating. As a result, our method can not only capture the updating time automatically, but also update the templates effectively. Furthermore, the effectiveness of the proposed method has been verified on a representative CNN-based and Transformer-based tracker, respectively. The experimental results on five popular benchmarks show that our tracker can achieve the state-of-the-art performance.
Förderung
- Guangxi Science and Technology
- National Natural Science Foundation of China
- Nature Science Foundation of Fujian Province
- Fundamental Research Founds for the Central Universities of Huaqiao University
- Guangxi “Bagui Scholar” Teams for Innovation and Research Project
- Guangxi Collaborative Innovation Center of Multi-Source Information Integration and Intelligent Processing
- Guangxi Talent Highland Project of Big Data Intelligence and Application
- Research Fund of Guangxi Key Laboratory of Multi-Source Information Mining & Security
Zitieren
APA
Gao, J., Zhong, B., & Chen, Y. (2023). Robust Tracking via Learning Model Update With Unsupervised Anomaly Detection Philosophy. IEEE Transactions on Circuits and Systems for Video Technology, 33(5), 2330-2341. https://doi.org/10.1109/tcsvt.2022.3221723
Vancouver
Gao J, Zhong B, Chen Y. Robust Tracking via Learning Model Update With Unsupervised Anomaly Detection Philosophy. IEEE Transactions on Circuits and Systems for Video Technology. 2023;33(5):2330-2341. doi:10.1109/tcsvt.2022.3221723
BibTeX
@article{Gao2023,
author = {Gao, Jie and Zhong, Bineng and Chen, Yan},
title = {Robust Tracking via Learning Model Update With Unsupervised Anomaly Detection Philosophy},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2023},
volume = {33},
number = {5},
pages = {2330--2341},
doi = {10.1109/tcsvt.2022.3221723}
}
EUVIMED