Λεπτομέρειες βιβλιογραφικής εγγραφής
| Τίτλος: |
Feature data analysis of dance movements by motion capture. |
| Συγγραφείς: |
Chao Lu |
| Πηγή: |
Journal of Measurements in Engineering; Sep2025, Vol. 13 Issue 3, p701-708, 8p |
| Θεματικοί όροι: |
Motion capture (Human mechanics), Kinect (Motion sensor), Quantitative research, Acquisition of data, Dance techniques, Convolutional neural networks, Data analysis |
| Περίληψη: |
Motion capture technology has been applied in more and more fields, but the research in the field of dance is relatively rare. In order to combine motion capture technology with dance research, better understand the characteristics of dance movements, and provide support for their digital analysis, this paper mainly studied the application of a motion capture technology called Kinect in the analysis of dance movement feature data. The skeleton data of different dance movements was first collected based on Kinect v2, and then the collected data was analyzed using a spatio-temporal graph convolutional network (ST-GCN). On the basis of the original ST-GCN, the multi-branch structure was adopted to realize co-occurrence feature learning, and the bone length feature and direction feature were introduced to further enrich the feature data. Experiments were carried out on the NTU RGB+D and dance datasets. It was found that the improved ST-GCN had better performance than other current motion classification approaches on the NTU RGB+D. The top-1 accuracy for cross-subject (CS) and cross-view (CV) was 92.4 % and 96.7 %, respectively, and the average accuracy of different dance movements for the dance dataset was 96.035. The findings confirm the effectiveness of the proposed approach in the analysis of dance movement feature data, and it can be applied in the actual research of dance movements. [ABSTRACT FROM AUTHOR] |
|
Copyright of Journal of Measurements in Engineering is the property of Extrica and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Βάση Δεδομένων: |
Complementary Index |