Motion Mining

  • Authors:
  • Stan Sclaroff;George Kollios;Margrit Betke;Romer Rosales

  • Affiliations:
  • -;-;-;-

  • Venue:
  • MDIC '01 Proceedings of the Second International Workshop on Multimedia Databases and Image Communication
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

A long-term research effort to support data mining applications for video databases of human motion is described. Due to the spatio-temporal nature of human motion data, novel methods for indexing and mining databases of time series data of human motion are required. Further, since data mining requires a significant sample size to accurately model patterns in the data, algorithms that automatically extract motion trajectories and time series data from video are required. A preliminary system for estimating human motion in video, as well as indexing and data mining of the resulting motion databases is described.