Taekwondo Action Recognition Method Based on Partial Perception Structure Graph Convolution Framework

Jianqiao Liang,Guocai Zuo,Baiyuan Ding

doi:10.1155/2022/1838468

Abstract

Action recognition in Taekwondo competitions and training is an important task, which can provide a very valuable reference factor for technicians, athletes, and coaches. We propose a graph convolution framework with part of the perception structure to recognize, decompose, and analyze Taekwondo actions. Taking advantage of the long short-term memory of a part of the perception structure, the recognized Taekwondo actions are marked in time series, and then features are extracted from the graph convolution level to obtain the spatial and temporal associations between joints. Predict the action category and perform score matching based on the manual tag database. Finally, it is verified on our self-made Taekwondo competition data set. Our method has an average accuracy of 90% in action recognition, and an average action score matching rate of 74.6%. The accuracy of action recognition is high, which provides great assistance to Taekwondo e training and competitions.

Highlights

With the rapid development of science and technology, the sports industry urgently needs the intervention of artificial intelligence technology. is field has gradually attracted a lot of research
Yan proposed the ST-graph convolutional network (GCN) method in 2018. is method uses the bone joint matrix as input to obtain the spatiotemporal features between the joint points. e method we propose takes the function of the module as the demarcation point, separately obtains the spatiotemporal features before the skeleton node, and maps into a dynamic network, abandoning the predefined graph
We examine the current state of work and research in the realm of action recognition before moving on to research on Taekwondo action recognition

Summary

Introduction

With the rapid development of science and technology, the sports industry urgently needs the intervention of artificial intelligence technology. is field has gradually attracted a lot of research. E development of these technologies has directly promoted the interpretation, prediction, and interaction of training actions in the sports industry. Different from the development of the scoring system, the model of athletes adapting to technical and tactical methods to achieve victory has gradually become standard [3]. E main focus of understanding the scoring actions and promoting victory is not unique to Taekwondo. It is important to understand the technical actions of players in taekwondo competitions. It has been the research object of other fighting sports such as karate [7], boxing [8], fencing [9], and judo [10]

Objectives

Methods

Findings

Conclusion