August 2, 2007

Adaptive Architectural Objects Perform Unique Choreography

Ruairi Glynn of Interactive Architecture blog writes about his latest project "Performative Ecologies" - his post and his project page.

In a nutshell, he has created adaptive objects with a light source and a twirling arm. Here are two pictures:

Each object has access to a database of stored gestural movements. (Watch short video and longer video to see examples of movements).

In performance mode, objects seek out human beings in the environment and then perform for them by pulling movements from their database. These objects are adaptive in the sense that they use facial recognition software to determine the level of interest of each audience member. The objects then select different stored choreography and combinations of this choreography to maximize attention levels of viewers. (They may also generate movements in real-time but I'm not sure).

Performing for an audience:

Using facial recognition software:

After each performance, the adaptive objects "convene." The objects with less success in the attention-grabbing arena acquire the "moves" of the more successful objects. Then there's another performance and the robots go back to work constantly modifying their choreography to have as wide an appeal as possible.

According Glynn, his project explores the nature of interactivity, not simply pre-programmed responsiveness, in architectural environments.

If I was to try and sum it up [his project] in a sentence, it is fundamentally about giving our architecture the ability to enter into dialog with us, rather than simply respond with fixed behaviours to fixed commands from us, to learm from its experiences and adapt its behaviours, to suggest new spatial configurations and see how we respond.

Very often I find that so called ‘interactive environments’ rarely enable the architecture to negotiate its behaviours, but rather follow pre-choreographed routines when triggered. More broadly, a great deal of misuse of the term ‘interactive’ is common in art, design and architectural discourse and I believe that this has diluted its true meaning and huge potential. [Read full post...]

Interactive or Responsive Dance Environments?

To what extent does Glynn's contention that the term "interactive environments" is often misused in art apply to similar explorations in the dance world? In other words, are some/many of the technology-based dance performances and installations truly interactive or are they primarily responsive?

Exploring the Nature of Dance

Following-up on Amanda's post this morning, "Dance and Film," on DCDanceblog, does this "Performative Ecologies" installation shed light on what makes the movement/dancing of human beings unique or not unique in the context of adaptive objects that can create, perform and share their own choreography?

I find it interesting that the adaptive quest of the objects in this project is completely driven by market forces. The only motivating question is how much attention am I getting and how do I get more of this attention. Could Glynn have used other objectives to drive the "creativity" of the adaptive objects? How about maximizing certain emotional reactions/states of human viewers? How about offending audiences? How about getting audiences to talk with each other? How about ignoring the humans all together and focusing on forming strong emotional ties with the other objects?

Vias taken:

- Networked Performance

- Rhizome.org

Posted by Doug Fox at 2:22 PM - Permalink | Comments (2) | TrackBacks (0)

June 7, 2006

A Dance Vocabulary for Performance Art

On the Networked Performance blog I came across an interesting write-up about a multimedia and performance art work by Ursula Endlicher.

For this installation, "Website Impersonations: The Amazons (.at versus .com)," Ursula creates a movement vocabulary that corresponds to the underlying HTML code of a webpage. You can watch a video that shows exactly how users interact with this art work (QuickTime video file is about 27 MB):

A user sits down on top of a giant trackball as they watch two side-by-side projection screens positioned in the corner of a room. One image is projected of a performer whose movements reflect the HTML code for Amazon.com while on the other screen the performer's movements reflect the HTML code for Amazon.at. The code is generated in real-time from a live Internet connection. By moving one's butt to roll the trackball, users can determine the position of the cursor and thus the corresponding movements that appear on screen. Here's a diagram of the installation set-up and a larger version of these technical requirements are available on Ursula's site:

I think that the online presentation of this installation would be enhanced if there were a separate movement vocabulary gallery that showed, in a visual fashion, the specific connections between HTML code and the individual movements. For example, what are the specific series of movements made by a performer when he or she encounters a tag for the beginning of an HTML table? By creating this video-based movement vocabulary library, both online and in-person audiences would have a better understanding of the inner workings of this installation and would likely enjoy it more.

Ursula Endlicher will be showing her work this Saturday, June 10th at 7:00 PM at 315 Broadway, 5th Floor (between Duane/ Thomas Street) in Lower Manhattan.

Posted by Doug Fox at 8:11 AM - Permalink | Comments (1) | TrackBacks (0)

March 13, 2006

Dancing Up a Storm with Gene Kelley

I enjoyed watching a mechanical umbrella performance done to "Singing in the Rain" designed by installation artist Peter William Holden (via Boing Boing).

Busby Berkeley choreographed dancers to mimic the motions of machines and modern inventions. “AutoGene” is the flipside of this. It’s a simple aesthetic looking robot composed of eight modified umbrellas mounted in a circular pattern. A cocktail of air hoses and electrical cables join these umbrellas to a central computer which enables “AutoGene” to produce a choreographed dance which erodes the machine's mechanical qualities.

And if you want more Singing in the Rain, watch this hip-hop dance take on this famous song in a Volkswagon TV commercial.

Posted by Doug Fox at 11:57 AM - Permalink | Comments (0) | TrackBacks (0)

September 28, 2005

Kick Ass Kung-Fu Embodied Gaming Experience

I just came across the website for the Kick Ass Kung-Fu immersive game experience. This large-scale real-life/virtual game combination is intriguing and parallels some of the developments in motion analysis and tracking that I wrote about yesterday when discussing the motion performance.

The videos on the home page and in the Gallery are fascinating to watch. A player starts by standing in a 5 meter cushioned playfield and then does battle against virtual enemies. A video camera captures the image of the player in real-time and inserts this image on two screens in a virtual environment. So as the player kicks, jumps and moves, these motions are tracked and represented in the virtual space.

From the Kick Ass Kung-Fu website, I can't get a good handle on how the real-time capture system works. It appears that players don't have to have any type of motion tracing devices connected to their bodies for the computer system to create a 3D image of the players.

Posted by Doug Fox at 6:09 PM - Permalink | Comments (1)