EMAT: Efficient feature fusion network for visual tracking via optimized multi-head attention

J Wang, C Lai, Y Wang, W Zhang - Neural Networks, 2024 - Elsevier
The tracking methods based on Transformer have shown great potential in visual tracking
and achieved significant tracking performance. The traditional transformer based feature …

Attention-driven memory network for online visual tracking

H Zhang, J Liang, J Zhang, T Zhang… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
A memory mechanism has attracted growing popularity in tracking tasks due to the ability of
learning long-term-dependent information. However, it is very challenging for existing …

Transformer tracking with multi-scale dual-attention

J Wang, C Lai, W Zhang, Y Wang, C Meng - Complex & Intelligent Systems, 2023 - Springer
Transformer-based trackers greatly improve tracking success rate and precision rate.
Attention mechanism in Transformer can fully explore the context information across …

A lightweight target tracking algorithm based on online correction for meta-learning

Y Qi, G Yin, Y Li, L Liu, Z Jiang - Journal of Visual Communication and …, 2024 - Elsevier
The traditional Siamese network based object tracking algorithms suffer from high
computational complexity, making them difficult to run on embedded devices. Moreover …