Superpixelizing Binary MRF for Image Labeling Problems
This provides a handy method for computer vision researchers to reduce computational cost in image labeling tasks, though it is incremental as it builds on existing superpixel and MRF techniques.
The paper tackles the lack of a principled approach to formulate Markov random field (MRF) models at the superpixel level by showing how a generic pixel-level binary MRF can be solved in superpixel space, resulting in a derived superpixel-level MRF energy that remains submodular.
Superpixels have become prevalent in computer vision. They have been used to achieve satisfactory performance at a significantly smaller computational cost for various tasks. People have also combined superpixels with Markov random field (MRF) models. However, it often takes additional effort to formulate MRF on superpixel-level, and to the best of our knowledge there exists no principled approach to obtain this formulation. In this paper, we show how generic pixel-level binary MRF model can be solved in the superpixel space. As the main contribution of this paper, we show that a superpixel-level MRF can be derived from the pixel-level MRF by substituting the superpixel representation of the pixelwise label into the original pixel-level MRF energy. The resultant superpixel-level MRF energy also remains submodular for a submodular pixel-level MRF. The derived formula hence gives us a handy way to formulate MRF energy in superpixel-level. In the experiments, we demonstrate the efficacy of our approach on several computer vision problems.