Tutorial: Graph-Cuts versus Level-Sets

European Conference on Computer Vision, Graz, Austria, May 6, 2006

Yuri Boykov, Daniel Cremers, Vladimir Kolmogorov


Among a multitude of image segmentation methods, the level set method and the graph minimal cut approaches have emerged as two powerful paradigms to compute the segmentation of images. Both methods are based on fundamentally different representations of images. Level sets are formulated as infinite-dimensional optimization on a spatially continuous image domain. The Graph Cuts, on the other hand, are defined as minimal cuts of a discrete graph representing the pixels of the image. In this tutorial, we want to review these two methods, show their strengths and limitations, and bridge the gap between these seemingly very different paradigms.


NOTE: Daniel, Vladimir, and I are very interested in popularizing the materials of this tutorial. Feel free to use our slides but please do not forget to acknowledge our work. We might also be interested in rerunning this seminar at your place,... especially if your place is by the blue sea or near beautiful mountains :)
Reruns of the tutorial:

- May 18, 2006, Oxford University, UK: joint seminar for Visual Geometry Group (Oxford University), Robotics Research Group (Oxford University), and Microsoft Research (Cambridge, UK).