Improving the Accuracy of Tracker by Linearized Transformer

Thang Dinh, Kien Trung, Thanh Nguyen Chi, Long Quoc

2023

Abstract

Visual object tracking seeks to correctly estimate the target’s bounding box, which is difficult due to occlusion, illumination variation, background clutters, and camera motion. Recently, Siamese-based approaches have demonstrated promising visual tracking capability. However, most modern Siamese-based methods compute target and search image features independently, then use correlation to acquire correlation information from two feature maps. The correlation operation is a straightforward fusion technique that considers the similarity between the template and the search region. This may be the limiting factor in the development of high-precision tracking algorithms. This research offers a Siamese refinement network for visual tracking that enhances and fuses template and search patch information directly without needing a correlation operation. This approach can boost any tracker performance and produces boxes without any postprocessing. Extensive experiments on visual tracking benchmarks such as VOT2018, UAV123, OTB100, and LaSOT with DiMP50 base tracker demonstrate that our method achieves state-of-the-art results. For example, on the VOT2018, LaSOT, and UAV123 test sets, our method obtains a significant improvement of 5.3% (EAO), 3.5% (AUC), and 2.9% (AUC) over the base tracker. Our network runs at approximately 30 FPS on GPU RTX 3070.

Paper Citation

in Harvard Style

Dinh T., Trung K., Nguyen Chi T. and Quoc L. (2023). Improving the Accuracy of Tracker by Linearized Transformer. In Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-626-2, pages 607-614. DOI: 10.5220/0011900900003411

in Bibtex Style

@conference{icpram23,
author={Thang Dinh and Kien Trung and Thanh Nguyen Chi and Long Quoc},
title={Improving the Accuracy of Tracker by Linearized Transformer},
booktitle={Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2023},
pages={607-614},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011900900003411},
isbn={978-989-758-626-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Improving the Accuracy of Tracker by Linearized Transformer
SN - 978-989-758-626-2
AU - Dinh T.
AU - Trung K.
AU - Nguyen Chi T.
AU - Quoc L.
PY - 2023
SP - 607
EP - 614
DO - 10.5220/0011900900003411