Unlock instant, AI-driven research and patent intelligence for your innovation.

A multi-target occlusion tracking method based on long-term and short-term prediction model

A predictive model, long-term and short-term technology, applied in image analysis, image enhancement, instruments, etc., can solve problems such as background occlusion, inter-target occlusion, missed detection, etc., to achieve the effect of improving accuracy, easy implementation, and improving engineering efficiency

Active Publication Date: 2022-04-15
WUHAN UNIV
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The appearance-based prediction model can solve the problem of missed detection and background occlusion, and the motion-based prediction model can solve the problem of object occlusion

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A multi-target occlusion tracking method based on long-term and short-term prediction model
  • A multi-target occlusion tracking method based on long-term and short-term prediction model
  • A multi-target occlusion tracking method based on long-term and short-term prediction model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] In order to facilitate those skilled in the art to understand and implement the present invention, the present invention will be described in further detail below in conjunction with the accompanying drawings. It should be understood that the implementation examples described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0046] The present invention provides an occlusion trajectory recovery method based on long-term and short-term prediction, which is different from most existing methods that can only deal with one occlusion type or one occlusion time length. Adaptively handle different occlusion durations and different occlusion types. The method of the present invention first designs an appearance-based prediction model and a motion-based prediction model to predict the position of the occluded target during the occlusion frame, and then designs an adaptive weighted fusion method to fuse the r...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-target occlusion tracking method based on a long-term and short-term prediction model, which includes: firstly designing a motion-based prediction model and an appearance-based prediction model to predict the position of the lost target; secondly designing an adaptive weighted fusion method to fuse The short-term prediction results are obtained from these two prediction results; then the position of the lost target is predicted by the spline interpolation method and the result is regarded as the long-term prediction result; finally, an adaptive weighted fusion method is designed to fuse the long-term and short-term prediction results. The method proposed by the present invention solves the problem of single processing of certain occlusion time in multi-target tracking, can effectively deal with different occlusion time lengths and different types of occlusion, effectively restores the position of the target during the occlusion frame, and improves the tracking efficiency Accuracy. In the quantitative and qualitative tests on the actual monitoring data set, the validity of the proposed method of the present invention is proved.

Description

technical field [0001] The invention relates to the technical field of monitoring target tracking, in particular to a multi-target occlusion tracking method based on a long-term and short-term prediction model. Background technique [0002] The task of multi-target tracking is to give a video and output the trajectories of all the targets appearing in the video. Multi-target tracking is widely used and can be applied to multimedia analysis, visual surveillance and sports analysis, etc. As an intermediate task in the field of computer vision, multi-target tracking has always been the focus of research scholars. [0003] The current mainstream multi-target tracking method is a detection-based tracking framework, that is, first using a detector to detect a target in a video, and then performing data association to obtain the final target trajectory. The detection results are used as input, and their quality has a huge impact on the tracking results. Although major breakthrou...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/246
CPCG06T7/248G06T2207/10016G06T2207/20084G06T2207/30241
Inventor 陈军孙志宏梁超阮威健陈金王晓芬王超
Owner WUHAN UNIV