Target tracking method based on attention mechanism and twin network and related equipment

Abstract

The invention provides a target tracking method based on an attention mechanism and a twin network, and related equipment, and the method comprises the steps: inputting a target template and a searchregion into a preset target tracking model, and outputting target tracking information of the target template in the search region through the preset target tracking model, wherein the target trackingmodel comprises a twin network, and a channel attention module and / or a space attention module are / is additionally arranged in the twin network. The embodiment of the invention provides the target tracking method and the related equipment, the channel attention module and / or the space attention module are / is added into the residual network, the average expected overlapping rate and robustness ofa twin tracking algorithm are remarkably improved, and the robustness of motion change, camera motion, shielding and size change attributes in tracking is improved, so that when the target tracking method provided by the embodiment of the invention is used for target tracking prediction, a relatively accurate result can be obtained.

Description

technical field [0001] The present invention relates to the technical field of terminal display control, in particular to a target tracking method and related equipment based on an attention mechanism and a twin network. Background technique [0002] Object tracking has always been a research hotspot in the field of computer vision, and it is applied to human-computer interaction, intelligent video surveillance, and traffic detection. The single target tracking is to give the bounding box of the target in the first frame of the tracking video, and then predict the bounding box of the target in the subsequent frames. The current target tracking method based on deep learning is a discriminative method, which is beginning to be ahead of the generative method. The target tracking algorithm based on the Siamese network trains the model end-to-end, by expressing target tracking as a cross-correlation problem. At present, the target tracking algorithm based on Siamese network has...

AI Technical Summary

Problems solved by technology

[0005] In view of the deficiencies in the above-mentioned prior art, the purpose of the present invention is to provide users with a target tracking method and related equipment based on attention mechanism and twin network, so as to overcome the problem that the tracking algorithm in the prior art is inefficient on multiple class attributes. Low stickiness, leading to the defect that the accuracy of target tracking results is unstable

Images