Laser & Optoelectronics Progress, Volume. 58, Issue 4, 0415004-1(2021)
Multimodel Integrated Siamese Network Visual Tracking
A siamese network visual tracking algorithm based on fusion multitask differentiated homogeneous models is proposed to improve the accuracy of the algorithm. First, the siamese network visual tracking and target segmentation models are fused in the decision-making layer. Then, they are combined with multiscale search area, contextual features, and multilearning rate model updating strategy to track. Different algorithms are evaluated using standard datasets, namely, VOT, OTB, LaSOT, and UAV123. Experimental results show that the proposed algorithm can stably track the object under the interference of occlusion, fast motion, and illumination change, among others.
Get Citation
Copy Citation Text
Manqiang Che, Shubin Li, Jinpeng Ge. Multimodel Integrated Siamese Network Visual Tracking[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0415004-1
Category: Machine Vision
Received: Jul. 28, 2020
Accepted: Aug. 12, 2020
Published Online: Mar. 19, 2021
The Author Email: Che Manqiang (1229462669@qq.com)