Region-based multi-feature matching speed-skating athlete multi-target tracking method

A multi-target tracking and athlete technology, applied in biometric recognition, image analysis, image enhancement, etc., can solve the problems of short tracking time and long game time, and achieve the effect of improving matching accuracy

Pending Publication Date: 2022-01-21
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In existing data sets MOT17, MOT19 and other data sets for pedestrians, pedestrians mostly move in a straight line, and the tracking time is relatively short
However, speed skaters compete for a long time and move according to the venue, so the sliding trajectory is relatively fixed, so it is impossible to perform a good match directly using the existing algorithm

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
  • Region-based multi-feature matching speed-skating athlete multi-target tracking method
  • Region-based multi-feature matching speed-skating athlete multi-target tracking method
  • Region-based multi-feature matching speed-skating athlete multi-target tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] Attached below figure 1 And the specific embodiment describes the present invention in further detail:

[0032] combine figure 1 As shown, an area-based multi-feature matching speed skater multi-target tracking method includes the following steps:

[0033] s1. Divide the current area according to the speed skating venue, and divide the area as attached figure 2 shown;

[0034] In order to improve the accuracy of detection and obtain better multi-target tracking results, the sports field is divided into 8 areas according to the movement characteristics of different speed skaters in the field, and simple straight line or curve tasks are performed between the areas.

[0035] s2. Detect the speed skater information in the picture according to the target detection algorithm;

[0036] In order to better apply various features to track and match speed skaters, we used various information of speed skaters in the picture in the target detection stage.

[0037] s21, the pos...

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 combines deep learning and a computer vision algorithm, and particularly discloses a region-based multi-feature matching speed-skating athlete multi-target tracking method, which comprises the following steps: s1, dividing a current region according to a speed skating field; s2, detecting speed skating athlete information in the picture according to a target detection algorithm; s3, carrying out regional division on the speed skating athlete information detected in the s2 according to the division regions in the s1; s4, matching the speed skating athletes in each region by using the multi-feature information; and s5, according to the tracking result of the speed skating athlete in each region, performing matching among the regions to form a final tracking result. According to the method, multi-target tracking of the speed-skating athletes is carried out more robustly in modes of region segmentation, multi-feature matching and the like according to feature information such as the competition duration and the speed skating field of the speed skating athletes.

Description

technical field [0001] The invention combines deep learning and computer vision algorithms, specifically disclosing a region-based multi-feature matching speed skater multi-target tracking method. Background technique [0002] Multi-target tracking is a key technology in the field of computer vision, and it is widely used in autonomous driving, intelligent monitoring, and behavior recognition. However, due to the complexity of the multi-target tracking task itself, it faces more challenges than target detection and single-target tracking tasks, such as target overlapping, drastic changes in appearance, and similar appearance. How to solve these problems more effectively is of great significance to the application of multi-target tracking technology, so in the past few decades, people have proposed a wide range of solutions. [0003] Multiple Object Tracking (Multiple Object Tracking, MOT) refers to the detection of pedestrians, cars, animals and other multiple targets in th...

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 Applications(China)
IPC IPC(8): G06V40/10G06V10/74G06K9/62G06T7/246G06T7/11
CPCG06T7/246G06T7/11G06T2207/30196G06F18/22
Inventor 李宗民王一璠孙奉钰
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products