面向民族动态艺术数字化的拉班舞谱自动生成

Automatic generation of Labanotation for national dynamic art digitalization

拉班舞谱是一种用于记录人体动作的符号记录体系,它是保护、继承和传播民族民间舞蹈及其他动作艺术的有力工具.目前,拉班舞谱的获取主要依靠专业人员手工记录,效率较低.本文基于三维人体运动捕捉数据,通过识别与拉班符号相对应的元素动作,提出了一种自动生成拉班舞谱的方法.方法的主要部分包括运动分割和元素动作分析.在运动分割中,利用速度阈值方法将人体运动捕捉数据分割成与舞谱符号对应的元素动作片段.然后,将动作片段与拉班舞谱的节奏相匹配,从而确保生成的舞谱符号更加规范和准确.在元素动作分析中,拉班舞谱将动作区分为支撑动作和非支撑动作,考虑到两类动作性质不同,因此本文采用两种方法分别对支撑与非支撑动作进行有针对性的分析.实验表明,本文的方法实现了将运动捕捉数据表示的舞蹈动作转换为数字化的拉班舞谱,并且自动生成的效率远高于手工记录.所以该方法可以快速记录濒临失传的民族舞蹈,为保护和传承动作类非物质文化遗产做贡献.

Labanotation is a symbol system for recording human motions. It is a powerful tool for protecting,inheriting and disseminating folk dances and other action arts. At present, Labanotation data are mainly recorded manually by professionals, and the working efficiency of the system is low. This paper proposes a method that automatically generates Labanotation from three-dimensional human-motion capture data. The method identifies the elemental movements corresponding to basic Laban symbols. The method contains two components:motion segmentation and unit movement analysis. The motion segmentation divides the motion capture data into segments that are easily identified by the speed threshold method. To standardize the Laban symbols and improve their accuracy, the motion segments are then matched to the Labanotation rhythms. In the unit movement analysis, Labanotation divides the movements into support and non-support ones. As the two movement types have different characteristics, they are identified by different methods. In experiments, the proposed method successfully transformed the movements represented by the motion-capture data into digitized Labanotation.The automatic Labanotation generation is far more efficient than manual recording. Therefore, the method can quickly record ethnic dances on the verge of loss, helping to protect and preserve the intangible cultural heritage of action arts.