Cross-camera multi-target tracking method and device, equipment and medium

Abstract

The invention discloses a cross-camera multi-target tracking method and device, equipment and a medium, and relates to the field of artificial intelligence, and the method comprises the steps: obtaining video frames shot by a plurality of cameras; determining a first type of targets which are located in an overlapped visual space region between different cameras and have the same shooting time in the video frame, and performing de-duplication processing on the first type of targets to obtain residual targets after de-duplication; and based on the time sequence, classifying the duplicate-removed residual targets shot at different shooting times and the second type of targets in the non-overlapping visual space region, so as to obtain path tracks corresponding to each duplicate-removed residual target and each second type of target in the non-overlapping visual space region. Therefore, matching of the target tracks in different cameras is not needed, the target deduplication and classification are carried out to obtain the cross-camera target tracks, and cross-camera multi-target tracking can be realized more accurately.

Description

technical field [0001] The present invention relates to the field of artificial intelligence, in particular to a method, device, device and medium for tracking multiple targets across cameras. Background technique [0002] Currently, object tracking is one of the most valuable research directions in the field of artificial intelligence machine vision. Usually, object tracking is divided into two subcategories: single object tracking (SOT, Single Object Tracking) and multi-object tracking (MOT, Multi Object Tracking): Single-target tracking focuses on the tracking of a specific target or the tracking problem of a simpler scene. At this time, the number of visible targets in the target area in the scene is very small; multi-target tracking is widely used and is often used in ordinary Simultaneous tracking of multiple targets in the scene. Currently, multi-target tracking problems are more involved, such as the autonomous driving dataset KITTI, which includes both vehicle and ...

AI Technical Summary

Problems solved by technology

[0003] For cross-camera target tracking algorithms, such as figure 1 As shown, for the target tracking of pedestrians, the existing methods usually use a two-step superposition method: the first step is to track the target on a single camera to form a local trajectory; the second step is to use the classic tracklet-to-tracklet matching The algorithm matches and stitches several output results of single-camera tracking; using this cross-camera target tracking method, a single segment tracking and then trajectory matching will cause performance degradation due to trajectory matching errors

Images