Multi-target tracking method and related device

Abstract

The invention discloses a multi-target tracking method and a related device. The multi-target tracking method comprises the steps: obtaining a first detection frame image, and calling M detection algorithms to carry out the processing of the first detection frame image, and obtaining W groups of detection targets; determining a first association relationship between the W groups of detection targets to generate first association information containing the first association relationship; acquiring a first tracking frame image to call M detection algorithms to process the first tracking frame image to obtain X groups of tracking targets; comparing the W groups of detection targets with the X groups of tracking targets to determine whether each group of detection targets in the W groups of detection targets is the same as the X groups of tracking targets; if not, determining a second association relationship between the X groups of tracking targets to generate second association information containing the second association relationship; and updating the first association information into second association information. By implementing the embodiment of the invention, a plurality of targets can be tracked continuously and accurately, and a stable tracking effect is ensured.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a multi-target tracking method and a related device. Background technique [0002] The current multi-object tracking (MOT) technology has become a relatively mature technology after decades of development, and has a very wide range of applications in various fields, such as missile tracking and crowd tracking. [0003] In daily life, a large number of cameras are often set up in public places for real-time monitoring, tracking pedestrians, pedestrians' belongings, vehicles, etc., to prevent abnormal phenomena and ensure the safety of ordinary residents. [0004] However, in the prior art, a unified tracking algorithm is used to track pedestrians, pedestrians' belongings, or vehicles, resulting in the inability to continuously and accurately track multiple targets, and the tracking effect is unstable. Contents of the invention [0005] Embodiments of the present invention pro...

AI Technical Summary

Problems solved by technology

[0004] However, in the prior art, when tracking pedestrians, pedestrians' belongings, or vehicles, a unified tracking algorithm is used for processing, resulting in the inability to continuously and accurately track multiple targets, and the tracking effect is unstable.

Images