Trajectory tracking method, system and device based on portrait recognition and storage medium

Abstract

The invention discloses a trajectory tracking method, system and device based on portrait recognition, a storage medium and a computer program product. The method comprises the steps of obtaining a real-time face image, and comparing the real-time face image with a preset reference face image to obtain a comparison result, judging whether the real-time face image is a target face image needing trajectory tracking or not according to the comparison result, and if yes, executing a portrait track search task based on the target face image to obtain a portrait track search result. According to the method, the real-time face image can be obtained, the behavior characteristics of the person corresponding to the face image can be analyzed, the action track of the person corresponding to the face image can be rapidly obtained, and the method can be used for tracking the track of a suspicious person.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to a trajectory tracking method, system, device and computer-readable storage medium based on portrait recognition. Background technique [0002] In recent years, the field of artificial intelligence has been fully developed, and it has been widely and flexibly applied to people's life or production activities. Among them, the application of portrait recognition technology in trajectory tracking for people's actions is particularly obvious. [0003] In the current portrait recognition technology, it is generally realized by capturing a person object through a camera, and the temperature, height, gender and other parameters of the person object can be captured, but there is a lack of feature analysis and behavior pattern for the captured person object. Analysis and other processes make the person recognition technology unable to be flexibly applied to trajectory track...

AI Technical Summary

Problems solved by technology

[0003] In the current portrait recognition technology, it is generally realized by capturing a person object through a camera, and the temperature, height, gender and other parameters of the person object can be captured, but there is a lack of feature analysis and behavior pattern for the captured person object. Analysis and other processes make the person recognition technology unable to be flexibly applied to trajectory tracking, which limits the development and application of person recognition for trajectory analysis.

Images