Optimization Simulation of Match between Technical Actions and Music of National Dance Based on Deep Learning

In the match between technical movements and music of folk dance, most important thing is to extract features effectively. DL algorithm one efficient methods video at present. this study, method applied matching optimization in dance. Using train corresponding relationship national given dance are adapted musical beat points. To better reflect degree correlation movement changes, change rate feature value used instead itself calculation. The genetic theory spatial skeleton timing diagram compared. experiment shows that optimized by can achieve 95.78% accuracy, synchronization reach 96.17%. Therefore, proposed study fully matches movements—music quality better. This expands a new perspective for research technology. It has certain practical theoretical significance.