Sunday, February 4, 2007

images from the dance

One of our failed attempts at tracing contours involved plotting specific points on the body (left hand, right hand, right shoulder, nose, etc) as they moved across the picture plane. This proved to be incoherent two-dimensionally but I still had hope for the potential of this kind of analysis. These images are the result of plotting paths for both hands, knees and feet (using the contours as underlays in both the "font" and "side" planes). By associating each control point on each path with one captured contour I was able to spatially represent velocity. This association also allowed for a more "timely" NURBS surface to be generated--control vertices intersect the edge of each NURBS at simultaneous positions along the movement path. In other words, the NURBS surface for the hands represents the space between the hands as they move in relation to each other simultaneously. The result is the condition of V or U vertices never being parallel (or perpendicular) to another (as would be the case if the geometrically simplest NURBS surface was generated between both paths).


maa said...

good work carl

ChenBuiYu said...

Awesome studies. I think you opened a whole avenue for investigation. You could go on and model hand to hand, foot to foot, knee to foot to head, and on. You could make a whole series of these and never run out!! (PS You got to show me how to fix the lighting in MAYA some time. The lighting in your models is really elegant.)