Adaptive Kernel Learning Kalman Filtering With Application to Model-Free Maneuvering Target Tracking

Li, Yuankai; Lou, Jiaxin; Tan, Xiaosu; Xu, Yan; Zhang, Jinpeng; Jing, Zhongliang

doi:10.1109/access.2022.3193101

Cited by 6 publications

(1 citation statement)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two main types of tracking techniques: Generative ones involve training the tracking model to represent the object appearance and then finding the object that best matches the object appearance [5]. This is usually done using particle [6], Kalman filter [7], or mean-shift [8]. Discrimination techniques, on the other hand, attempts to separate the target from the surrounding environment to achieve higher accuracy using DCF-based trackers [9].…”

Section: Introductionmentioning

confidence: 99%

Leveraging Attention for Improved Discriminative Correlation Filters in Single Object Tracking

Elsaid,

Fouad,

Ghoniemy

2024

Preprint

View full text Add to dashboard Cite

Correlation Filter-based Trackers have shown impressive results in the object tracking research, outperforming classical trackers in several benchmarks. However, accurately tracking objects with deformation, fast motion, or occlusion remains a main challenge in the process of tracking. The cyclic suggestion of training samples used in correlation filter tracking usually lead to undesirable boundary effects, that significantly reduce the tracking efficiency. To address this issues, a hybrid attention-based correlation filter method (Att-DCF) is proposed for robust object tracking. By developing a discrimination filter and reducing the limitations of irrelevant features, the proposed approach provides more accurate and consistent estimated target locations. The proposed Att-DCF tracker is evaluated using the recent and standard datasets, OTB-100, Temple-Colour128, and UAV123. Compared to the baseline DCF tracker, the proposed tracker achieves an improvement of 2.71$\%$ and 1.38$\%$ in terms of area under the curve and precision measures using the OTB-100 dataset, 3.99$\%$ and 3.43$\%$ using the Temple-Colour128, and 4.72$\%$ and 1.76$\%$ using the UAV123, respectively.

show abstract