Frank R. Schmidt

Hausdorff Distance Constraint for Multi-Surface Segmentation

European Conference on Computer Vision (ECCV), Volume 7572, page 598--611 - Oct 2012
Download the publication: 707 KB 
Share the publication:
It is well known that multi-surface segmentation can be cast as a multi-labeling problem. Different segments may belong to the same semantic object which may impose various inter-segment constraints. In medical applications, there are a lot of scenarios where upper bounds on the Hausdorff distances between subsequent surfaces are known. We show that incorporating this prior into multi-surface segmentation is potentially NP-hard. To cope with this problem we develop a submodular-supermodular procedure that converges to a locally optimal solution well-approximating the problem. While we cannot guarantee global optimality, only feasible solutions are considered during the optimization process. Empirically, we get useful solutions for many challenging medical applications including MRI and ultrasound images.

Images and movies

brain_minmax_4.png (99 KB)

BibTex references

  author       = "Schmidt, Frank R. and Boykov, Yuri",
  title        = "Hausdorff Distance Constraint for Multi-Surface Segmentation",
  booktitle    = "European Conference on Computer Vision (ECCV)",
  series       = "LNCS",
  volume       = "7572",
  pages        = "598--611",
  month        = "Oct",
  year         = "2012",
  publisher    = "Springer",
  address      = "Florence, Italy",
  url          = ""