A Labanotation Editing Tool for Description and Reproduction of Stylized Traditional Dance Body Motion

poster / demo / art installation
  1. 1. Worawat Choensawat

    School of Science and Engineering - Ritsumeikan University

  2. 2. Sachie Takahashi

    School of Letters - Ritsumeikan University

  3. 3. Minako Nakamura

    Ochanomizu University

  4. 4. Kozaburo Hachimura

    School of Science and Engineering - Ritsumeikan University

Work text
This plain text was ingested for the purpose of full-text search, not to preserve original formatting or readability. For the most complete copy, refer to the original conference program.

A Labanotation Editing Tool for Description and Reproduction of Stylized Traditional Dance Body Motion
Choensawat, Worawat, School of Science and Engineering, Ritsumeikan University , gr0011es@ed.ritsumei.ac.jp
Takahashi, Sachie, School of Letters, Ritsumeikan University , nr013082@ed.ritsumei.ac.jp,
Nakamura, Minako, Graduate School of Humanities and Sciences, Ochanomizu University , nakamuraminako@gmail.com
Hachimura, Kozaburo, School of Science and Engineering, Ritsumeikan University , hachimura@media.ritsumei.ac.jp
A stylized traditional dance has uniqueness in itself which reflects history, culture, emotion expression, etc. When recording and representing this traditional dance body motions, it is important to have capabilities for handling these very characteristic body movements, which can probably be handled with the full-set of Labanotation (Hutchinson, 1997). However, notation score becomes extremely complicated, and we will not be able to comprehend what is that movement.

We are facing a problem how to realize a method of describing detailed features and nuance of artistic, traditional dance movements while suppressing the complexity in notation score.

Several graphics applications has been developed for preparing Labanotation scores (Brown and Smoliar, 1976; Fox, 2000) and generating body movement (Calvert, 2007; Coyle et al. 2005; Coyle et al. 2002; Wilke et al. 2005). However none of them solved the abovementioned problem yet.

We have been working on a system named LabanEditor (Kojima et al. 2002; Nakamura and Hachimura, 2006). It includes functionalities of both inputting/editing Labanotation score and displaying character animation so that beginners who are not familiar with Labanotation can study about its description by a trial-and-error approach.

In this paper, we aim at description and reproduction of the body motion of stylized traditional dances by using fundamental elements of Labanotation while keeping the quality of body movement of CG character animation. We propose and implement a dynamic template technique enabling users to notate stylized traditional dances and reproducing it in 3D CG animation from a Labanotation score.

Labanotation is a graphical notation system for recording human body movements. Figure 1 (a) is an example of Labanotation scores corresponding to dance motion. A Labanotation score is drawn in a form of vertical staff where each column corresponds to a body part. Figure 1 (b) shows the basic arrangement of columns in the staff. The horizontal dimension of the staff represents the parts of the body, and the vertical dimension represents time.

Figure 1. Labanotation scores; (a) Example of Labanotation scores and (b) Columns of Labanotation representing body parts.

Full Size Image

LabanEditor (Kojima et al. 2002) is an interactive graphical editor for editing Labanotation scores and displaying the 3D CG character animation associated with scores.

We added the new features to LabanEditor as follow:

Dynamic template technique enabling users to notate movements and reproducing it in 3D CG animation using fundamental description of Labnotation.
Motion control module for manipulate the motion expression among key-frames in order to make the animation more natural.
User Interface
With LabanEditor, users are able to input and edit the score and then display the CG animation immediately, which makes possible to interactively confirm the movements. Users can zoom in/out and change the viewpoint of the 3D scene on all three axes.

While replaying the Labanotation score, users can observe the animation as well as the red horizontal line cursor, moving upward corresponding to the animation progresses, as shown in Figure 2.

Figure 2. CG animation display window.

Full Size Image

Generating the CG Animation
Labanotation scores can be represented as a simple format called LND (Kojima et al. 2002), which uses alphanumeric characters to represent basic symbols. To create 3D character animation, we have to convert LND into animation data. The format of LND representation is shown in Figure 3. The lines that begin with “#” indicate the fundamental parameters of Labanotation. The movement of a body part is specified by the line followed by a command “direction”, which corresponds to the Labanotation direction symbols.

Figure 3. Example of LND.

Full Size Image

LND describes a pose just like key-frame body postures for animation, so that we can produce motion of the body part by simply applying interpolation between start and end key-frame poses. A key-frame pose of a body part at a timing corresponding to an end of the symbol is defined by a Labanotation symbol.

Our system converts direction symbols into animation key-frames by using a template file for a mapping between the symbol and its corresponding pose of the body part.

The template file describes the relationship between a direction symbol at the particular column and the rotation and the translation of the corresponding joint. Figure 4 shows a notation and description in a template file, and the resulting pose.

Figure 4. Relationship between user input symbols and a template file; (a) User input symbol, (b) Part of a template file, and (c) Snapshot of the CG animation corresponding to the template in (b).

Full Size Image

Dynamic Template Technique
Due to the rough resolution of fundamental elements of Labanotation, similar but distinct poses are sometimes defined with the same symbol in a Labanotation score.

To solve this problem, we invented the method of dynamic templates in order to represent very specific movements using the fundamental subset of Labanotation symbols only. With the dynamic template technique, we can represent these characteristic motions by changing the template files dynamically during a display process, while using very fundamental symbols.

Figure 5 (a) shows the interface of editing template file. Users can activate the editing window by double clicking on a Labanotation symbol. For example, suppose the symbol in the Labanotation score, indicated by a red color in Figure 5 (a), was selected by a user, then, the user can directly edit the joint angles on an editable template panel as shown in Figure 5(a).

Alternatively, the graphically editable template, which is activated by clicking the „Animate‟ button in Figure 5 (a), enables the user to edit the template by adjusting the slide bars and observe the resulting pose as shown in Figure 5 (b).

Figure 5. Interface for editing a template; (a) Template editing panel and (b) Graphical template editing panel.

Full Size Image

Full Size Image

The information of template files is inserted into a LND file corresponding to the start time as shown in Figure 6. The command “#include” determines the template file used at a particular timing. As a result, in this case, the Labanotation score shown in Figure 5(a) will be interpreted as the LND file shown in Figure 6.

Figure 6. Example of LND file using dynamic templates.

Full Size Image

During the animation process, the Labanotation symbols, in format of LND, are mapped to the key-frame pose indicated by the current template.

Motion Expression Control
The motion expression control module controls the animation of character model from one key-frame to the next key-frame. We implemented a module for controlling the motion by applying a non-linear interpolation, cubic Bezier curve, in order to create natural movement.


Where f(t) is an interpolated position or joint angle at time t and a normalized time scaled from 0.0 to 1.0, respectively. P0(0,0) and P3(1,1) is the start and end points, respectively, while P1 and P2 are the control points which can be moved freely as shown in Figure 7.

Figure 7. Motion expression controller user interface.

Full Size Image

Figure 8 shows two snapshots of the CG animation corresponding to the motion expression graphs on the left.

Figure 8. CG animation snapshots corresponding to the motion expression graphs.

Full Size Image

Use of LabanEditor for Noh Plays
Noh is the most famous and characteristic Japanese traditional performing arts in the stylized form of a musical dance-drama that has been performed since the 14th century. There are about 240 plays in the repertoire from five Noh schools (Ortolani, 1995).

Noh body movement is Japanese highly stylized and is not the same as ordinary human movement. Therefore, the direction symbols used in Labanotation must be interpreted differently when we handle Noh plays and generate body motion from it. This has been realized by preparing motion template files which are editable to represent specific motions in that particular performance.

Snapshots of Noh motion are shown in Figure 9. Figure 9 (a) and (b) show the reproduced animation of Noh body motion from Labanotation score using the Noh templates and standard templates, respectively.

Figure 9. CG character animation of Noh Kata (Shitai) (a) CG character animation corresponding to the Labanotation score on the left using Noh (Shitai) template (b) CG character animation corresponding to the Labanotation score on the left using standard template.

Full Size Image

In this paper we proposed an approach of description and reproduction of stylized traditional dances such as Noh plays with Labanotation. A new version of LabanEditor, LabanEditor3, successfully describes and reproduces Noh motions by using the dynamic template method.

We have obtained a major achievement. Our approach shows that we can describe and reproduce Noh plays with the fundamental description of Labanotation, with a limited number of symbols by using the dynamic template method.

Hutchinson, A 1977 “Labanotation, ” Theatre Arts Books

Brown, M. D. Smoliar, S.W. 1976 “A Graphics Editor for Labanotation, ” Proceedings of the 3rd Annual Conference on Computer Graphics, Interactive Techniques and Image Processing, ACM Computer Graphics, 10 (2) 60-65

Fox, I 2000 “Documentation Technology for the 21st Century, ” World Dance 2000 Academic Conference, Papers and Abstracts, 137–142

Calvert, T 2007 “Animating Dance, ” Proceedings of Graphics Interface, 1-2

Calvert, T Wilke , L Ryman, R and Fox, I 2005 “Applications of Computers to Dance, ” IEEE Computer Graphics Application, 25 2 6-12

Coyle, M Maranan, D and Calvert, T 2002 “A Tool for Translating Dance Notation to Animation, ” Proceedings of Western Computer Graphics Symposium,

Wilke, L Calvert, T Ryman, R & Fox, I 2005 “From dance notation to human animation, The LabanDancer Project, Motion Capture and Retrieval, ” Computer Animation and Virtual Worlds, 16 3-4 201-211

Kojima, K Hachimura, K & Nakamura, M 2002 “LabanEditor: Graphical Editor for Dance Notation, ” Proceedings of IEEE 2002 International, Workshop on Robot and Human Interactive Communication, 59-64

Nakamura, M and Hachimura, K 2006 “An XML Representation of Labanotation, LabanXML, and Its Implementation on the Notation Editor LabanEditor2, ” . Review of the National Center for Digitization (Online Journal), 9 47-51

Ortolani, B 1995 The Japanese Theatre: From Shamanistic Ritual to Contemporary Pluralism, Princeton University Press

If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.

Conference Info


ADHO - 2011
"Big Tent Digital Humanities"

Hosted at Stanford University

Stanford, California, United States

June 19, 2011 - June 22, 2011

151 works by 361 authors indexed

XML available from https://github.com/elliewix/DHAnalysis (still needs to be added)

Conference website: https://dh2011.stanford.edu/

Series: ADHO (6)

Organizers: ADHO

  • Keywords: None
  • Language: English
  • Topics: None