Daniel Melancon


A kinect-based interactive visual installation that will transform each viewer into a different particle-based visualization.


Introduction to Computational Media

Transfigure is a interactive visual installation. A Kinect is mounted above a large screen/tv that is capturing depth and RGB data of the area in front of the screen as if a large mirror. When the user enters into the Kinect's field of vision, their user data will be transformed into one of six visual characterizations. Taking in up to six viewers at once, each new viewer will be randomly assigned from any of the possible six characterizations that are not already taken. The visualizations are made up of different particle systems created by 2D points being drawn in a 3D space as well as some static depth-based visualizations. There is an microphone input on the laptop as well that is capturing the audio level around the screen. When there a noise over a certain threshold, like a hand clap, every viewers visualization will randomize. The particle system-based characters will dissipate and the new visualization will emerge. Because only the viewers body is captured, each viewer sees themselves against a stark white background for a visually appealing, provocative and immersive experience.