 |
|
|
 |
Home Information Registration Sessions Schedule Participants Contacts |
SESSIONS Laban Motif of Algorithms for Automatic Generation of Dance Sequences in "Web3D Dance Composer" by Bin Umino, Japan, Jeffrey Scott Longstaff, UK, and Asako Soga, Japan Paper Recent discussions regarding usage of Labanotation signs (ICKL technical sessions, Beijing 2004) have led to the wider question of the relation between methods of motif description compared to full structured Labanotation description. This research offers one perspective on the application of Laban motif for use in documenting movement events. “Web3D Dance Composer” (http://pinakes.soc.toyo.ac.jp/BIN/webdance/ ) consists of an online archive and user editable simulation system for ballet steps and step sequences. An exhaustive archive of 3D motion captured ballet movements can be browsed, previewed, and selected to compose extended sequences. In addition the automatic composing system allows ballet steps to be combined automatically according to an algorithm which describes a characteristic style of the dance sequence. The system is available free on the internet, with minimum system requirements of a Windows PC (Windows 95/98/Me/NT/2000/XP), Microsoft Internet Explorer, and a plug-in VRML viewer such as ‘Cortona VRML Client’ available for free download. The user interface allows a variety of manipulations (Fig. 1). The VRML viewer displays the 3D motion captured dance movement allowing the view point of the performing area to be rotated and panned. The ‘Display control panel’ allows manipulations of type of dancer, type of stage space, number of dancers, and arrangement of dancers. The ‘Step catalogue panel’ allows selection of positions, steps, and entire pre-composed sequences. The ‘Motion control panel’ allows selection of duration of steps and speed of performance. This panel also allows steps to be added to the composed sequence, rewound, played, paused, and manually moved forward or backwards with the horizontal scroll bar. Using the ‘Time line panel’, users can select ballet steps one by one, arrange them into a sequence, and combine all steps to make an entire composition. Using the ‘Recording panel’, users can save their original sequences in plain text code. In addition, by using the ‘Auto’ button, the automatic composing system self-generates an entire dance sequence according to a particular algorithm prescribing its structure. Originally designed for generation of petit allegro enchaînement, further variations on this algorithm have also been derived for generation of new original dance sequences such as might occur in contemporary choreographies. The algorithms used by Web3D Dance Composer system in the automatic generation of dance sequences consist of collections of rules, choices, and constraints according to which the motion captured ballet movements in the archive are selected and combined into dance sequences. Each subsequent composition by the same algorithm generates a new unique sequence. In an attempt to gain a perceptual understanding of an algorithm, Laban motif would offer an appropriate method for documentation, allowing the entire collection of algorithmic rules, options, and choices to be represented in a visuo-graphic display and according to a language of body movement developed from the Laban analysis system (Fig. 2). Laban motifs of algorithms were included in the online user interface of Web3D Dance Composer. These were organised into a hierarchical structure to reveal three layers of organisation. In some cases specialised motif symbols were derived in order to express the pattern of choices available in the algorithm. Further, the online motif display was linked to a catalogue giving resources, definitions and usage for all motif signs used. By documenting an algorithm for generating movement sequences in Laban motif, these mathematical models can be expressed in a language of body movement and made more readily perceivable. Motif methods also give an example of how improvisational or non-linear compositions might be documented by representing dance sections as collections of underlying rules and options. The Web3D Dance Composer is valuable for online virtual dance experimentation and exploration by teachers and choreographers involved in creative practices, improvisation, creative movement, or dance composition. It is also be useful for teachers and students of Labanotation and Laban motif as one example of how the graphic signs can be arranged and applied in documentation of a movement event.
Bin Umino is a Professor at the Department of Media and Communication of Toyo University in Tokyo. He is lecturing some courses in Informatics at the university. He graduated doctoral course in Library and Information Science at University of Tokyo in 1991. He has been leading "Web3D Dance Project" for educational and artistic purposes since 1999. Jeffrey Scott Longstaff (MS, CMA, PhD) consults on movement research and is principal editor of Laban-analyses <http://www.laban-analyses.org>, an online searchable database of practitioners and annotated bibliography for works utilising Laban-based methods. Asako Soga is a Research Associate at the Department of Media Informatics, Faculty of Science and Technology, Ryukoku University. She received her BE in Engineering from Ibaraki University in 1999, her ME in Human Informatics from Nagoya University in 2001, and her PhD in Human Informatics from Nagoya University in 2004. Her research interests are computer animation, networked virtual environments, and their applications for artistic media. Currently she is working on human animation systems for dance using computer, the Internet, and motion captured data.
|
