4.6 Article

Robust cascaded-parallel visual tracking using collaborative color and correlation filter models

Journal

MULTIMEDIA TOOLS AND APPLICATIONS
Volume -, Issue -, Pages -

Publisher

SPRINGER
DOI: 10.1007/s11042-023-15614-4

Keywords

Correlation filter; Cascaded-parallel tracking; Target likelihood probability; Reliable expert selection analysis

Ask authors/readers for more resources

In recent years, the multi-expert collaborative tracking strategy has been introduced into visual tracking tasks and achieves impressive performance. Different from most existing multi-expert trackers, the proposed cascaded-parallel tracking algorithm (CPT) adaptively selects the suitable expert among multiple tracking models. With the proposed adaptive expert selection mechanism, the most reliable expert (i.e. tracking model) is selected for tracking in each frame. Extensive experimental results demonstrate that our proposed algorithm performs favorably against some state-of-the-art algorithms on various datasets.
In recent years, the multi-expert collaborative tracking strategy has been introduced into visual tracking tasks and achieves impressive performance. Different from most existing multi-expert trackers that linearly fuse multiple tracking models, we propose a novel cascaded-parallel tracking algorithm (CPT) via adaptively selecting the suitable expert among multiple tracking models. And the CPT consists of cascaded and parallel tracking components. In the cascaded tracking component, we hierarchically implement two effective correlation filter models to coarse-to-fine locate the target. And in the parallel tracking component, a color tracking model is applied to locate the target to compensate for the demerit of the correlation filter models. With the proposed adaptive expert selection mechanism, the most reliable expert (i.e. tracking model) is selected for tracking in each frame. Extensive experimental results on OTB2013, OTB2015 and TempleColor128 datasets demonstrate that our proposed algorithm performs favorably against some state-of-the-art algorithms.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available