Thesis Proposal MoviesHere are some QuickTime 4 movies that link from my thesis proposal:
Swamped! (27.1MB/audio) Shows the various aspects of Swamped!.
Michael Patrick Johnson narrates. (6.4MB/audio) A
subsequence of the above video that shows control and gesture
recognition.
Luxomatic (6.0MB/no audio)
Shows the Luxomatic system. The
actual system runs much faster than shown here. But you get the idea.
PHMM (0.7MB/no audio)
An animation depicting a parametric hidden Markov model with a
one-dimensional parameterization on a two handed gesture indicating a
quantity; as in "I caught a fish; it was this big." The parameter
is the size of the fish.
Seagull (7.0MB/no audio) The
Seagull performance animation
system.
Natural Gesture
(11.6MB/audio) 1. Annotated sequence: A clip of the original video used in
the natural gesture coding system experiments. Red annotations are
determined manually, blue annotations are derived by the system. R = rest
state, T = transition, S = stroke.
(6.9MB/audio) 2. Coding Example: Fixed rate sequence. The same sequence
showing only every 10th frame.
(6.8MB/audio) 3. Coding Example: Variable rate, manually annotated
sequence. The same number of frames, chosen with respect to manual
annotation to code triphasic (iconic) gestures richly.
(6.8MB/audio) 4. Coding Example: Variable rate, automatically annotated
sequence. The same, but now with automatic selection of frames derived from
automatic gesture classification.
Watch and Learn
(5.1MB/audio) View from the camera while Watch and
Learn learns a simple conducting gesture.
(7.8MB/audio) GUI view from the same session, showing appearance models,
gamma traces (blue) and prior probabilities from the score (red).
Andy Wilson, awilson@media.mit.edu Last modified: Fri Jun 11 14:54:01 EDT 1999 |