Isoperimetric Graph Partitioning for Image Segmentation

  • Authors:
  • Leo Grady;Eric L. Schwartz

  • Affiliations:
  • IEEE;-

  • Venue:
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Year:
  • 2006

Quantified Score

Hi-index 0.14

Visualization

Abstract

Spectral graph partitioning provides a powerful approach to image segmentation. We introduce an alternate idea that finds partitions with a small isoperimetric constant, requiring solution to a linear system rather than an eigenvector problem. This approach produces the high quality segmentations of spectral methods, but with improved speed and stability.