Deep Learning-Based Action Recognition Algorithm for Jiangsu Dan Opera Characters
Jiabei Li, Yu Xin · WSEAS TRANSACTIONS ON COMPUTER RESEARCH · 2025
With the rapid development of artificial intelligence technology, deep learning has achieved remarkable results in fields such as image recognition. This study takes Jiangsu Dan opera, an opera genre with unique regional cultural characteristics, as the object, and conducts an in-depth analysis of the artistic characteristics and character movements of Dan opera based on deep learning algorithms to clarify the importance and difficulties of movement recognition. By collecting a large amount of Dan opera image and video data, a dataset suitable for Dan opera character action recognition is constructed. Advanced deep learning models, such as deep learning ST-GCN network and OpenPose, a multi-person pose estimation algorithm, are used to train and optimize the dataset. The experimental results show that the proposed algorithm has high accuracy and robustness in Dan opera character movement recognition, can effectively identify different movement types, and provides a new technical means for the inheritance, protection, and innovative development of Dan opera.