Automatic Generation of Labanotation Based On Extreme Learning Machine with Skeleton Topology Feature
Xueyan Zhang, Zhenjiang Miao, Qiang Zhang · 2018
Labanotation is one of the most widely used notation systems for recording and analyzing human movement. In this paper, we propose a novel method for automatic generation of Labanotation based on skeleton topology feature and Extreme Learning Machine. Firstly, according to the principle of human body organization structure, we design the feature that represent the characteristics of human motion better. Then we adopt the efficient and fast Extreme Learning Machine algorithm to recognize human movement. Finally, we draw the Labanotation score according to the identified movement categories corresponding to the symbols of Labanotation. This is the first time that a neural network based approach has been used in the research of automatically generating Labanotation. Experimental results show that the proposed approach achieves much better performance than the sate-of-the-art methods in recognition accuracy and computational time, demonstrating the efficiency of our method.