DoSer is intended to be a proof-of-concept implementation of an image segmentation algorithm which uses a graph-theoretical approach, namely the notion of dominant sets to cluster images.
DoSer is based on the work of Massimiliano Pavan and Marcello Pelillo, titled Efficiently Segmenting Images with Dominant Sets. See the full paper at Springer.
See the milestones for an estimated development plan. See the wiki for additional details.