Siamese Network for Fast Visual Tracking of Rotating Targets

Yibo Gao

doi:10.23977/vcip.2022.010101

Siamese Network for Fast Visual Tracking of Rotating Targets

Download as PDF

DOI: 10.23977/vcip.2022.010101 | Downloads: 12 | Views: 1580

Author(s)

Yibo Gao ¹

Affiliation(s)

¹ Guidance Control and Information Perception Laboratory of High Overload Ammunition, Army Artillery and Air Defense Academy of PLA, Hefei, Anhui, 230031, China

Corresponding Author

Yibo Gao

ABSTRACT

Directional target tracking is an important task in the field of target tracking, which has great application prospects in geography, agriculture and military. The current algorithm for rotating target detection relies on the detection frame after regression and then uses the segmentation mask for further accuracy, which is obviously too cumbersome. In this paper, a scheme of direct generation of directional detection frame (Siamese-ORPN) is proposed. Specifically, improve the alternative box strategy so that the orpn can directly generate a high-quality directional detection box proposal in a low consumption manner. In addition, a top-down feature fusion network is proposed as the backbone of feature extraction and feature fusion, which can obtain substantial benefits from the diversity of visual semantic levels. Siamese-ORPN realizes lightweight and real-time detection, and achieves leading performance on benchmark data sets, including vot2018 (44.6% EAO) and vot2019 (39.6% EAO).

KEYWORDS

Oriented target tracking, feature extraction, RPN, Feature fusion

CITE THIS PAPER

Yibo Gao, Siamese Network for Fast Visual Tracking of Rotating Targets. Visual Communications and Image Processing (2022) Vol. 1: 1-7. DOI: http://dx.doi.org/10.23977/vcip.2022.010101.

REFERENCES

[1] A. W. Smeulders, D. M. Chu, R. Cucchiara, S. Calderara, A. Dehghan, and M. Shah. Visual tracking: An experimental survey. IEEE Trans. Pattern Anal. Mach. Intell., vol. 36, no. 7, pp. 1442–1468, Jul. 2014.
[2] M. Danelljan, G. Bhat, F. S. Khan, and M. Felsberg. (2017) ECO: Effificient convolution operators for tracking.in Proc. IEEE Conf. Comput. Vis.Pattern Recognit, pp. 6638–6646.
[3] M. Danelljan, G. Bhat, F. S. Khan, and M. Felsberg. (2019) ATOM: Accurate tracking by overlap maximization.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 4660–4669.
[4] K. Dai, D. Wang, H. Lu, C. Sun, and J. Li. (2019) Visual tracking via adaptive spatially-regularized correlation fifilters.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 4670–4679.
[5] L. Bertinetto, J. Valmadre, J. F. Henriques, A. Vedaldi, and P. H. S. Torr. (2016) Fully-convolutional Siamese networks for object tracking.in Proc. Eur. Conf. Comput. Vis. Workshop, pp. 850–865.
[6] B. Li, W. Wu, Q. Wang, F. Zhang, J. Xing, and J. Yan. (2019) SiamRPN++: Evolution of Siamese visual tracking with very deep networks.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 4282–4291.
[7] Z. Chen, B. Zhong, G. Li, S. Zhang, and R. Ji. (2020) Siamese box adaptive network for visual tracking.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 6668–6677.
[8] R. Yao, G. Lin, S. Xia, J. Zhao, and Y. Zhou. (2020) Video object segmentation and tracking: A survey. ACM Trans. Intell. Syst. Technol., vol. 11, no. 4, pp. 1–47.
[9] Q. Wang, L. Zhang, L. Bertinetto, W. Hu, and P. H. S. Torr. (2019) Fast online object tracking and segmentation: A unifying approach.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 1328–1338.
[10] B. Chen and J. K. Tsotsos. (2019) Fast visual object tracking with rotated bounding boxes, arXiv: 1907. 03892.
[11] A. Lukežiˇc, J. Matas, and M. Kristan. (2020) D3S–A discriminative single shot segmentation tracker. in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 7133–7142.
[12] J. Ma et al. (2018) Arbitrary-oriented scene text detection via rotation proposals. IEEE Trans. Multimedia, vol. 20, no. 11, pp. 3111–3122, Nov.
[13] J. Ding, N. Xue, Y. Long, G. Xia, and Q. Lu. (2019) Learning RoI transformer for oriented object detection in aerial images.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 2849–2858.
[14] X. Xie, G. Cheng, J. Wang, X. Yao, and J. Han. (2021) Oriented R-CNN for object detection.in Proc. IEEE Int. Conf. Comput. Vis, pp. 3520–3529.
[15] L. Bertinetto, J. F. Henriques, J. Valmadre, P. H. S. Torr, and A. (2016) Vedaldi. Learning feed-forward one-shot learners, arXiv: 1606. 05233.
[16] T. Y. Lin et al. (2014) Microsoft COCO: Common objects in context.in Proc. Eur. Conf. Comput. Vis, pp. 101–115.
[17] O. Russakovsky et al. (2015) ImageNet large scale visual recognition challenge. Int. J. Comput. Vis., vol. 115, pp. 211–252.
[18] M. Kristan et al. (2018) The sixth visual object tracking VOT2018 challenge results.in Proc. Euro. Conf. Comput. Vis. Workshops, pp. 3–53.
[19] M. Kristan et al. (2019) The seventh visual object tracking VOT2019 challenge results.in Proc. IEEE/CVF Int. Conf. Comput. Vis, pp. 2206–2241.
[20] G. Wang, C. Luo, Z. Xiong, and W. Zeng. (2019) SPM-tracker: Series-parallel matching for real-time visual object tracking.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 3643–3652.
[21] Y. Xu, Z. Wang, Z. Li, Y. Yuan, and G. Yu. (2020) SiamFC++: Towards robust and accurate visual tracking with target estimation guidelines.in Proc. AAAI Conf. Artif. Intell, pp. 12549–12556.
[22] G. Bhat, M. Danelljan, L. V. Gool, and R. (2019) Timofte. Learning discriminative model prediction for tracking.in Proc. IEEE Int. Conf. Comput. Vis. pp. 6182–6191.
[23] Z. Fu, Q. Liu, Z. Fu, and Y. Wang. (2021) “STMtrack: Template-free visual tracking with space-time memory networks. In Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 13774–13783.
[24] S. Cheng et al. (2021) Learning to fifilter: Siamese relation network for robust tracking.in Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp. 4421–4431.

Subscription

E-Mail Alert

Downloads:	62
Visits:	4053

Siamese Network for Fast Visual Tracking of Rotating Targets

Author(s)

Affiliation(s)

Corresponding Author

ABSTRACT

KEYWORDS

CITE THIS PAPER

REFERENCES

RESOURCES

JOIN US

PUBLICATION SERVICES

CONTACT US