Unsupervised Image Segmentation by Oriented Image Foresting Transform in Layered Graphs

In this work, we address the problem of unsupervised image segmentation, subject to high-level constraints expected for the objects of interest. More specifically, we handle the segmentation of a hierarchy of objects with nested boundaries, each with its own expected boundary polarity constraint. To this end, this work successfully extends Hierarchical Layered Oriented Image Foresting Transform (HLOIFT), with the inclusion of nested object relations, to the unsupervised segmentation paradigm. On the other hand, this work can also be seen as an extension of Unsupervised OIFT (UOIFT) to include structural relationships of nested objects. The method is demonstrated in the segmentation of three datasets of colored images with superior performance compared to other existing techniques in graphs, requiring a smaller number of connected partitions to isolate the objects of interest in the images.