Similarity measures for image matching despite occlusions in stereo vision

  • Authors:
  • Sylvie Chambon;Alain Crouzil

  • Affiliations:
  • IFSTTAR, MACS, AI, Centre de Nantes route de Bouaye, CS 05 44 344, 44341 Bouguenais Cedex, France and UPS, IRIT, TCI, 118 route de Narbonne, 31062 Toulouse Cedex 9, France;UPS, IRIT, TCI, 118 route de Narbonne, 31062 Toulouse Cedex 9, France

  • Venue:
  • Pattern Recognition
  • Year:
  • 2011

Quantified Score

Hi-index 0.01

Visualization

Abstract

In the context of computer vision, matching can be done with similarity measures. This paper presents the classification of these measures into five families. In addition, 18 measures based on robust statistics, previously proposed in order to deal with the problem of occlusions, are studied and compared to the state of the art. A new evaluation protocol and new analyses are proposed and the results highlight the most efficient measures, first, near occlusions, the smooth median powered deviation, and second, near discontinuities, a non-parametric transform-based measure, CENSUS.