Target tracking method based on segmentation target mask updating template

Abstract

The invention discloses a target tracking method based on a segmentation target mask updating template. The target tracking method comprises the following steps: constructing a basic network frameworkof target tracking; initializing the network, obtaining foreground information in the regression frame, generating an initialization target template with a highlighted foreground, and linearly superposing the initialization target template with the initialization target template; inputting the linear superposition result into a target template branch of a tracking module, and obtaining the central position and the size of a next frame of target; at intervals of m frames, enabling the network to calculate a target template corresponding to the frame through a target center point of the corresponding frame and regression frame parameters, and input the target template of the frame into the mask segmentation module to generate a subsequent frame target template of the highlighted foreground;linearly superposing the initialization target template, the initialization target template with the highlighted foreground and the subsequent frame target template with the highlighted foreground togenerate a new target template used by the next frame; and inputting the target template into a tracking network framework, and calculating the central position and the size of the next frame of target.

Description

technical field [0001] The invention relates to the field of deep neural networks, in particular to a target tracking method based on segmented target mask update templates under the deep twin network framework. Background technique [0002] With the increasing development of artificial intelligence, computer vision is more and more widely used in various fields and people's daily life. As an important branch in the field of computer vision, target tracking technology plays a pivotal role in different application fields such as automatic driving, human-computer interaction, pedestrian detection, and weapon precision strikes. It also has very broad application prospects and far-reaching implications. Significance. [0003] The basic task of target tracking can be briefly summarized as follows: Given the size and position of the target in the first frame of the video, the algorithm calculates the center position and size of the target frame by frame in subsequent frames, ther...

AI Technical Summary

Problems solved by technology

This type of algorithm usually uses a large-scale data set to train the model end-to-end and track it offline. Therefore, it lacks an online update strategy for adjusting the filter template similar to the correlation filtering algorithm. In the application scenarios, the stability of the tracking algorithm will decline to varying degrees

Images