The axes are the first and the second principal components based on several features calculated from the motion data. The pies show the distribution of the verbs used for describing the videos. The size of a pie is relative to the amount of participant that described the video. Click on a pie to see the related video. Click on a colored box to toggle the color on/off. For more details read the related paper Impact of Varying Vocabularies on Controlling Motion of a Virtual Actor.