Robust interactive multi-label segmentation with an advanced edge detector

Recent advances on convex relaxation methods allow for a flexible formulation of many interactive multi-label segmentation methods. The building blocks are a likelihood specified for each pixel and each label, and a penalty for the boundary length of each segment. While many sophisticated likelihood estimations based on various statistical measures have been investigated, the boundary length is usually measured in a metric induced by simple image gradients. We show that complementing these methods with recent advances of edge detectors yields an immense quality improvement. A remarkable feature of the proposed method is the ability to correct some erroneous labels, when computer generated initial labels are considered. This allows us to improve state-of-the-art methods for motion segmentation in videos by 5-10% with respect to the F-measure (Dice score).

@inproceedings{MOWG16,
  title        = {Robust interactive multi-label segmentation with an advanced edge detector},
  author       = {S. M{\"u}ller and P. Ochs and J. Weickert and N. Graf},
  year         = {2016},
  editor       = {B. Andres and B. Rosenhahn},
  booktitle    = {German Conference on Pattern Recognition (GCPR)},
  series       = {Lecture Notes in Computer Science},
  publisher    = {Springer},
  volume       = {9796},
  pages        = {117--128}
}