MGMM: Multiresolution Gaussian Mixture Models for Computer Vision

  • Authors:
  • Affiliations:
  • Venue:
  • ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 1
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper introduces a new generalization of scale-space and pyramids, which combines statistical modeling with a spatial representation. The representation uses the familiar concept of multiple resolutions, but applied to a Gaussian mixture representation of the image - hence the title MGMM. It is shown that MGMM can approximate any probability density. Examples show how MGMM can be applied to problems such as segmentation and motion analysis.