By nature motion is inherently time-based. In a mediated context this does not have to be the case. The temporal dimension of motion can directly be mapped onto a spatial one. The result is an entirely static object that inherits the dynamic properties from the temporal axis.

For this piece we invited dancers to the NYU motion capture lab. We recorded sequences of jumps, spins, and rolls. With this data we rendered still frames of abstracted human figures which we fed into isosurf. Isosurf considers these frames as cross-sections of a volume and generates its triangulated surface. In this series of images the x-axis corresponds to the time-axis of the original movment. The two main features of the shape are a crouch and a jump.

There are three discreet steps when going from mocap data to rendered shape. First we saved screenshots of the mocap visualization with this c/opengl function. Then we concatenated all the images with this shell command. Finally we ran isosurf with these options.

1.) opengl raw image 2.) concatenate images 3.) isosurf

back to → fabricating information