Target tracking method and device, electronic equipment and computer readable storage medium

Abstract

The embodiment of the invention provides a target tracking method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining an ROI (regionof interest) in an Nth frame of image, wherein the ROI in the Nth frame of image comprises an image of a target, and N is a positive integer; and performing target tracking on the (N + 1) th frame ofimage by using the Nth frame of image and the ROI in the Nth frame of image. According to the embodiment of the invention, a region of interest ROI in an Nth image is acquired and the target trackingmode is utilized to perform target tracking on the ROI in the (N + 1) th frame of image, and since the target tracking mode is that a certain association exists between two adjacent frames of images,the stability and continuity of the ROI tracking result are ensured.

Description

technical field [0001] The present application relates to the field of image processing, and in particular, to a method, apparatus, electronic device and computer-readable storage medium for target tracking. Background technique [0002] Object detection is a key technology in video processing and computer vision. [0003] In order to improve the accuracy of target detection, the prior art discloses a real-time video detection target algorithm. The video target algorithm continuously improves the utilization of image features through the continuous improvement of the low-level convolutional layer and the high-level convolutional layer. The predicted image is up-sampled to obtain a feature layer with a larger size, and the obtained up-sampled feature layer is channel-connected to the lower-level network, and a second layer of prediction is added to make the algorithm detect the target more accurately. [0004] Although the real-time video detection target algorithm uses two ...

AI Technical Summary

Problems solved by technology

[0004] Although the real-time video detection target algorithm improves the detection accuracy by adopting two-layer prediction, each frame of image processed by the video detection target algorithm is independent, and there is no necessary connection between frames, thus It may appear that the same target exists in consecutive N frames of images, and N is a positive integer greater than 1, but the detection result may be that a part of the frame images detects the target, and another part of the frame images does not detect the target, thus causing the same target in Discontinuity in the video, that is, the existing video detection target algorithm has at least the problem that the detection continuity of the same target in consecutive frame images in the video cannot be guaranteed

Images