Efficiently Locating Objects Using the Hausdorff Distance

  • Authors:
  • William J. Rucklidge

  • Affiliations:
  • Xerox Palo Alto Research Center, 3333 Coyote Hill Road, Palo Alto, CA 94304. E-mail: rucklidge@parc.xerox.com

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Hausdorff distance is a measure defined between two point sets,here representing a model and an image. The Hausdorff distance isreliable even when the image contains multiple objects, noise,spurious features, and occlusions. In the past, it has been used tosearch images for instances of a model that has been translated, ortranslated and scaled, by finding transformations that bring a largenumber of model features close to image features, and vice versa. Inthis paper, we apply it to the task of locating an affinetransformation of a model in an image; this corresponds todetermining the pose of a planar object that has undergoneweak-perspective projection. We develop a rasterised approach to thesearch and a number of techniques that allow us to locate quickly alltransformations of the model that satisfy two quality criteria; wecan also efficiently locate only the best transformation. We discussan implementation of this approach, and present some examples of itsuse.