Discovering Clusters in Motion Time-Series Data (Preprint)

Abstract

A new approach is proposed for clustering time-series data. The approach can be used to discover groupings of similar object motions that were observed in a video collection. A finite mixture of hidden Markov models (HMMs) is fitted to the motion data using the expectation-maximization (EM) framework. Previous approaches for HMM-based clustering employ a k-means formulation, where each sequence is assigned to only a single HMM. In contrast, the formulation presented in this paper allows each sequence to belong to more than a single HMM with some probability, and the hard decision about the sequence class membership can be deferred until a later time when such a decision is required. Experiments with simulated data demonstrate the benefit of using this EM-based approach when there is more overlap in the processes generating the data. Experiments with real data show the promising potential of HMM-based motion clustering in a number of applications.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 26, 2003
Accession Number
ADA461872

Entities

People

  • George Kollios
  • Jonathan Alon
  • Stan Sclaroff
  • Vladimir Pavlovic

Organizations

  • Boston University

Tags

Communities of Interest

  • Autonomy
  • Counter IED
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Aspect Ratio
  • Classification
  • Computational Science
  • Computer Science
  • Computers
  • Databases
  • Hidden Markov Models
  • Information Science
  • Machine Learning
  • Markov Models
  • Models
  • Network Science
  • Probabilistic Models
  • Probability
  • Probability Distributions

Fields of Study

  • Computer science

Readers

  • Image Processing and Computer Vision.
  • Regression Analysis.
  • Speech Processing/Speech Recognition.