Data Visualization with Dance
The core idea is to visualize statistical properties of
single and multi-variate data by means of dance movements of computer generated
animation figures.
Traditional visualizations utilize basic dimensions of
graphical representation to portray multi-modal time-varying data: color,
shape, size, and location. The dimensionality of these representations is
limited and their aesthetic quality on the whole is generic. Moreover, the
visualization is often static, the time dimension is
"frozen-out". This IDF aims to
make improvements to the visual display of time-varying data.
The first computerized dance notation system, which
displayed an animated figure on the screen which performed the dance moves
specified by the choreographer, was the DOM dance notation system, created by
Eddie Dombrower on the Apple II personal computer in 1982. (See Dance Notation
Journal, Fall, 1986, 4(2) pp. 47-48.)
This solution consisted of a single figure, and the movements of that
figure were not based on any external data sources or their statistical
properties.
There are numerous COTS packages used in Web Design and
Electronic Games industry that enable one to create figures and to animate and
display those figures. I envision
leveraging such tools in building this system.
I propose there to use the language of dance in particular
and the language of movement in general to visualize statistical properties of
single and multi-variate time varying data.
These streams of data may originate from sensors, from real-time
structured and un-structured data, or may be the outputs of forecasting
calculations.
My approach consists of several steps. Through a user interface, a user selects the
data stream(s) that he wishes to visualize using dance. Next, he uses software
based tools to create a figure the movements of whose limbs are going to
indicate the data and its statistical properties. For each data stream he will create a
distinct figure. The figures will differ
in color, shape, and size thus visually indicating the distinct time-varying
data streams that are being visualized.
The system saves the results of the figure creation.
Next, the user, will choreograph these figures based on
statistical properties of the data. For
example, the user may decide to assign specific dance moves to those figures
for which the data streams are beyond a certain threshold as defined by the
mean-value of the data, or as defined by degrees from the standard deviation of
that mean. Or the user may decide to
indicate those data streams that have positive correlations with one another by
figures that dance with and around each other in close proximity of one
another- depending on the degree of the correlation. In a similar manner, the user may choreograph
these figures with dance steps so that other statistical properties of data are
thereby indicated - higher moments of the distribution function and so on.
The user will input the choreography in the form of one of
the available dance notation systems such as Labanotation system or the Sutton
Movement system. An embodiment of this
invention using the Sutton Movement system enables inclusion of skate-boarding,
pantomime, gymnastics and other such activities as ways of indicating
time-varying data. The user connects his
dance choreography with the statistical properties of the time-varying data
through this interface.
The system saves the results of the choreography which
consists of dance movements as well as statistical properties that trigger
those movements and guide them.
At this stage, the user has accomplished 3 tasks: he has
created figures, he has identified his data streams with specific figures, and
he has choreographed dance moves for each figure based on the statistical
properties of (potentially all of) these time-varying data streams.
Next the user indicates to the system to animate these
figures based on the input data streams and the dance moves that were
choreographed and saved earlier. The
system will begin processing the data stream(s) and compute the statistical
properties of the time-varying data.
Based on these statistical properties, the system will automatically load
the figure from its data store and invoke the corresponding dance movements for
each figure (data stream). The system
will then display animated dancing figures on a display device that indicate
the statistical properties of the data stream(s).
The display device and the delivery of the images is not
part of this; those task scan be accomplished through WEB, Client-Server,
Mobile, or other architectures and technologies.
Web References:
Journal Articles:
- Eddie Dombrower, Dance Notation Journal, Fall, 1986, 4(2) pp. 47-48.
- M. Cunningham, Changes/Notes on Choreography, F. Starr, ed., Something Else Press, 1968.
- D. Tolani, A. Goswami, and N. Badler, "Real-Time Inverse Kinematics Techniques for Anthropomorphic Limbs," Graphical Models, vol. 62, no. 5, 2000, pp. 353-388; .
- M. Van de Panne, "From Footprints to Animation," Computer Graphics Forum, vol. 16, no. 4, 1997, pp. 211-223.
- L. Wilke et al., "Animating the Dance Archives," Proc. 4th Int'l Symp. Virtual Reality, Archaeology and Intelligent Cultural Heritage (VAST), Eurographics Assoc., 2003, pp. 91-99.
- M. Nakamura, "Text Representation of Labanotation Data for Computer Based Motion Analysis," presented at the World Dance Assoc./Int'l Council of Kinetography Laban/Congress on Research in Dance Int'l Conf., 2004; (http://www.imb.is.ritsumei.ac.jp/~hachihachi_e.html)
- A. Hutchinson Guest, Labanotation: The System of Analyzing and Recording Movement, Taylor and Francis, 1987.
- Tom Calvert , Lars Wilke, Rhonda Ryman, Ilene Fox, “Applications of Computers to Dance”, IEEE Computer Graphics and Applications March/April 2005 (Vol. 25, No. 2) pp. 6-12.
- “A Prototype Program for Visualization of Dance Performances Using 3DCG Motion Database”,
- Umino Bin(Fac. of Socil., Toyo Univ.) Soga Asako (Ryukoku Univ. Fac. Sci. and Technol.)
- IPSJ SIG Technical Reports, ISSN:0919-6072, VOL.2006;NO.85(CH-71);PAGE.41-46(2006)
Books:
Patent:
“Dance visualization
of music”, United States Patent 6717042
Abstract:
An
apparatus is equipped to provide dance visualization of a stream of music. The
apparatus is equipped with a sampler to generate characteristic data for a
plurality of samples of a received stream of music, and an analyzer to
determine a music type for the stream of music using the generated
characteristic data. The apparatus is further provided with a player to
manifest a plurality of dance movements for the stream of music in accordance
with the determined music type of the stream of music.
No comments:
Post a Comment