The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching
Abstract This work aims to explore an accurate and effective method for recognizing dance movement features, providing precise personalized guidance for sports dance teaching. First, a human skeletal graph is constructed. A graph convolutional network (GCN) is employed to extract features from the n...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-04-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-96462-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850184286626381824 |
|---|---|
| author | Yi Xie Yao Yan Yuwei Li |
| author_facet | Yi Xie Yao Yan Yuwei Li |
| author_sort | Yi Xie |
| collection | DOAJ |
| description | Abstract This work aims to explore an accurate and effective method for recognizing dance movement features, providing precise personalized guidance for sports dance teaching. First, a human skeletal graph is constructed. A graph convolutional network (GCN) is employed to extract features from the nodes (joints) and edges (bone connections) in the graph structure, capturing both spatial relationships and temporal dynamics between joints. The GCN generates effective motion representations by aggregating the features of each node and its neighboring nodes. A dance movement recognition model combining GCN and a Siamese neural network (SNN) is proposed. The GCN module is responsible for extracting spatial features from the skeletal graph, while the SNN module evaluates the similarity between different skeletal sequences by comparing their features. The SNN employs a twin network structure, where two identical and parameter-sharing feature extraction networks process two input samples and calculate their distance or similarity in a high-dimensional feature space. The model is trained and validated on the COCO dataset. The results show that the proposed GCN-SNN model achieves an accuracy of 96.72% and an F1 score of 86.55%, significantly outperforming other comparison models. This work not only provides an efficient and intelligent personalized guidance method for sports dance teaching but also opens new avenues for the application of artificial intelligence in the education sector. |
| format | Article |
| id | doaj-art-eeac733ca413409f9f8a180e2c6a55c5 |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-eeac733ca413409f9f8a180e2c6a55c52025-08-20T02:17:05ZengNature PortfolioScientific Reports2045-23222025-04-0115111410.1038/s41598-025-96462-0The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teachingYi Xie0Yao Yan1Yuwei Li2Xinyu UniversityBeijing Institute of Graphic CommunicationChina Civil Affairs UniversityAbstract This work aims to explore an accurate and effective method for recognizing dance movement features, providing precise personalized guidance for sports dance teaching. First, a human skeletal graph is constructed. A graph convolutional network (GCN) is employed to extract features from the nodes (joints) and edges (bone connections) in the graph structure, capturing both spatial relationships and temporal dynamics between joints. The GCN generates effective motion representations by aggregating the features of each node and its neighboring nodes. A dance movement recognition model combining GCN and a Siamese neural network (SNN) is proposed. The GCN module is responsible for extracting spatial features from the skeletal graph, while the SNN module evaluates the similarity between different skeletal sequences by comparing their features. The SNN employs a twin network structure, where two identical and parameter-sharing feature extraction networks process two input samples and calculate their distance or similarity in a high-dimensional feature space. The model is trained and validated on the COCO dataset. The results show that the proposed GCN-SNN model achieves an accuracy of 96.72% and an F1 score of 86.55%, significantly outperforming other comparison models. This work not only provides an efficient and intelligent personalized guidance method for sports dance teaching but also opens new avenues for the application of artificial intelligence in the education sector.https://doi.org/10.1038/s41598-025-96462-0Artificial intelligenceSiamese neural networkSports dance teachingPersonalized guidance |
| spellingShingle | Yi Xie Yao Yan Yuwei Li The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching Scientific Reports Artificial intelligence Siamese neural network Sports dance teaching Personalized guidance |
| title | The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching |
| title_full | The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching |
| title_fullStr | The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching |
| title_full_unstemmed | The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching |
| title_short | The use of artificial intelligence-based Siamese neural network in personalized guidance for sports dance teaching |
| title_sort | use of artificial intelligence based siamese neural network in personalized guidance for sports dance teaching |
| topic | Artificial intelligence Siamese neural network Sports dance teaching Personalized guidance |
| url | https://doi.org/10.1038/s41598-025-96462-0 |
| work_keys_str_mv | AT yixie theuseofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching AT yaoyan theuseofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching AT yuweili theuseofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching AT yixie useofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching AT yaoyan useofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching AT yuweili useofartificialintelligencebasedsiameseneuralnetworkinpersonalizedguidanceforsportsdanceteaching |