Abstract- The presentation of a robust framework for automatically classifying Punjabi folk dance poses using pose estimation techniques and machine learning algorithms. Through the integration of advanced deep learning models for pose estimation and meticulously designed feature extraction methods, the research achieves high classification accuracy across a diverse array of poses. The successful classification of Punjabi folk dance poses holds significant implications for the preservation and promotion of cultural heritage, facilitation of dance education and choreography, and enhancement of human-computer interaction in dance performance analysis. While the study opens avenues for future research, including exploration of the approach’s generalization to other dance forms and cultural contexts, and efforts to enhance scalability and efficiency for real-time applications, it contributes to the advancement of computer vision and pattern recognition research. By showcasing the efficacy of combining pose estimation and machine learning techniques for fine-grained action recognition tasks in cultural domains, this work adds to the growing body of literature in both computer vision and dance research communities. Additionally, this section provides an extensive review of pertinent literature on pose estimation, action recognition, and cultural dance analysis, focusing on methodologies and techniques applicable to the classification of Punjabi folk dance poses.