Image Segmentation Using Minimal Graph Cuts
Anders P. Eriksson, Olof Barr, Kalle Åström · QUT ePrints (Queensland University of Technology) · 2006
We present a novel technique for simultaneous segmentaion and classification of image partitions using graph cuts. By combining existing image segmentation approaches with simple learning techniques we manage to include prior knowledge into this visual grouping process. This has resulted in a method that partitions images into two parts based on previously seen example segmentations. Prelimininary results are also presented in support of our suggested approach.