Video data processing method and device

Abstract

Embodiments of the present invention provide a video data processing method and apparatus. The video data processing method includes the steps of acquiring sample video data collected by a plurality of video collection apparatuses respectively, wherein the sample video data collected by one of the plurality of video collection apparatuses and the sample video data collected by at least one another video collection apparatuses of the plurality of video collection apparatuses record a motion trajectory of the same sample target; and training with the acquired sample video data so as to determine a location relation among the plurality of video collection apparatuses. According to the video data processing method and apparatus provided by the embodiments of the present invention, the location relation among the plurality of video collection apparatuses can be determined through a method of training with the samples, and the method is simple to implement, and has low requirements on deployment of the video collection apparatuses and prior information such as a background scene and so on, so that the deployment overhead of a target track system can be reduced.

Description

technical field [0001] The present invention relates to the technical field of target tracking, and more particularly relates to a video data processing method and device. Background technique [0002] Location tracking technology plays an important role in enterprise security and other fields. The traditional tracking method is mainly based on a single camera. The advantage of this method is that it has lower requirements for the shooting scene and camera layout, and it is more convenient in the deployment stage of the tracking system, but the disadvantage lies in the tracking of a specific target (such as a specific person). It takes a lot of manpower and time, because all video data needs to be checked one by one manually. The recently proposed multi-camera tracking method can automatically track a specific target among multiple cameras by using the positional relationship of the cameras and background information. However, this method often requires relatively harsh in...

AI Technical Summary

Problems solved by technology

However, this method often requires relatively harsh initialization conditions, such as the need to perform 3D modeling of the tracking area in advance, and the need to reasonably arrange the distribution of the camera, etc.

Images