Cargando…
Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets
Issues of fairness and consistency in Taekwondo poomsae evaluation have often occurred due to the lack of an objective evaluation method. This study proposes a three-dimensional (3D) convolutional neural network–based action recognition model for an objective evaluation of Taekwondo poomsae. The mod...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10575175/ https://www.ncbi.nlm.nih.gov/pubmed/37836879 http://dx.doi.org/10.3390/s23198049 |
_version_ | 1785120865180975104 |
---|---|
author | Luo, Chenglong Kim, Sung-Woo Park, Hun-Young Lim, Kiwon Jung, Hoeryong |
author_facet | Luo, Chenglong Kim, Sung-Woo Park, Hun-Young Lim, Kiwon Jung, Hoeryong |
author_sort | Luo, Chenglong |
collection | PubMed |
description | Issues of fairness and consistency in Taekwondo poomsae evaluation have often occurred due to the lack of an objective evaluation method. This study proposes a three-dimensional (3D) convolutional neural network–based action recognition model for an objective evaluation of Taekwondo poomsae. The model exhibits robust recognition performance regardless of variations in the viewpoints by reducing the discrepancy between the training and test images. It uses 3D skeletons of poomsae unit actions collected using a full-body motion-capture suit to generate synthesized two-dimensional (2D) skeletons from desired viewpoints. The 2D skeletons obtained from diverse viewpoints form the training dataset, on which the model is trained to ensure consistent recognition performance regardless of the viewpoint. The performance of the model was evaluated against various test datasets, including projected 2D skeletons and RGB images captured from diverse viewpoints. Comparison of the performance of the proposed model with those of previously reported action recognition models demonstrated the superiority of the proposed model, underscoring its effectiveness in recognizing and classifying Taekwondo poomsae actions. |
format | Online Article Text |
id | pubmed-10575175 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-105751752023-10-14 Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets Luo, Chenglong Kim, Sung-Woo Park, Hun-Young Lim, Kiwon Jung, Hoeryong Sensors (Basel) Article Issues of fairness and consistency in Taekwondo poomsae evaluation have often occurred due to the lack of an objective evaluation method. This study proposes a three-dimensional (3D) convolutional neural network–based action recognition model for an objective evaluation of Taekwondo poomsae. The model exhibits robust recognition performance regardless of variations in the viewpoints by reducing the discrepancy between the training and test images. It uses 3D skeletons of poomsae unit actions collected using a full-body motion-capture suit to generate synthesized two-dimensional (2D) skeletons from desired viewpoints. The 2D skeletons obtained from diverse viewpoints form the training dataset, on which the model is trained to ensure consistent recognition performance regardless of the viewpoint. The performance of the model was evaluated against various test datasets, including projected 2D skeletons and RGB images captured from diverse viewpoints. Comparison of the performance of the proposed model with those of previously reported action recognition models demonstrated the superiority of the proposed model, underscoring its effectiveness in recognizing and classifying Taekwondo poomsae actions. MDPI 2023-09-23 /pmc/articles/PMC10575175/ /pubmed/37836879 http://dx.doi.org/10.3390/s23198049 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Luo, Chenglong Kim, Sung-Woo Park, Hun-Young Lim, Kiwon Jung, Hoeryong Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title | Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title_full | Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title_fullStr | Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title_full_unstemmed | Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title_short | Viewpoint-Agnostic Taekwondo Action Recognition Using Synthesized Two-Dimensional Skeletal Datasets |
title_sort | viewpoint-agnostic taekwondo action recognition using synthesized two-dimensional skeletal datasets |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10575175/ https://www.ncbi.nlm.nih.gov/pubmed/37836879 http://dx.doi.org/10.3390/s23198049 |
work_keys_str_mv | AT luochenglong viewpointagnostictaekwondoactionrecognitionusingsynthesizedtwodimensionalskeletaldatasets AT kimsungwoo viewpointagnostictaekwondoactionrecognitionusingsynthesizedtwodimensionalskeletaldatasets AT parkhunyoung viewpointagnostictaekwondoactionrecognitionusingsynthesizedtwodimensionalskeletaldatasets AT limkiwon viewpointagnostictaekwondoactionrecognitionusingsynthesizedtwodimensionalskeletaldatasets AT junghoeryong viewpointagnostictaekwondoactionrecognitionusingsynthesizedtwodimensionalskeletaldatasets |