Here’s one of my earliest experiments with light-based hyperinstruments, in this case colored bottle caps, and a very simple form of computer vision algorithm. Pandora (2006) is the first performance-centric project I completed at Virginia Tech and can be best described as a demo or a technological etude, if you like, rather than an art piece, where I explored MaxMSP‘s Jitter library and its potential for a reasonably accurate computer vision. The end result is a simple hyperinstrument built from LEDs placed inside colored bottle caps with straps and battery packs that can be used to control real-time granular synthesis of a sampled snippet of my own voice.

I remember designing a vocabulary of approximately a dozen different gestures, some of which rely on a relative position of both hands, and others being hand-specific, something I found being particularly exciting. In addition, raw x and y position data was used to change timbre, as well as spatialize the ensuing audio data across quad audio output. The real-time video feed from the embedded webcam was used both for computer vision and as a processed texture projected onto a virtual cube whose abstract visuals served more as a visual instrument feedback to the performer, than an artistic visualization.

While in and of itself a mere milestone, Pandora in some ways marks a turning point in my career, where I finally began acknowledging and openly pursuing my passion for transdisciplinary creativity, something that has since played a pivotal role in clarifying and defining my professional identity and aspirations.

Performances:

• DISIS Digital Id 070429 happening in Blacksburg, Virginia, April 2007. (premiere)
• Oberlin College Festival, Ohio, May 2007.