This page gives an example of using the kinematic segmentations to
find important postures of the arm during everyday activities.
Clustering performed on the hand positions at the detected transitions
between action segments (local minima in hand speed), results in
clusters that correspond with significant positions of the hand during
everyday activity. The following images show three important hand
positions discovered by the system that correspond well with the hand
being at rest by the wearer's side, the hand reaching out into the
world, and the hand carrying something. The top of each of the three
images shows the four discovered hand states as spheres, with the red
sphere being the state classification for that particular example. The
middle and bottom of the images show a typical captured image and
kinematic posture for that hand state.