Segmentation hi\'erarchique faiblement supervis\'ee

TL;DR

This paper introduces a versatile hierarchical image segmentation method that incorporates prior spatial information to emphasize regions of interest, demonstrating effectiveness in weakly-supervised segmentation tasks.

Contribution

It presents a novel hierarchical segmentation approach that integrates prior spatial data, enhancing focus on important structures in weakly-supervised settings.

Findings

Effective in emphasizing contours and regions of interest

Preserves important image structures

Applicable to weakly-supervised segmentation

Abstract

Image segmentation is the process of partitioning an image into a set of meaningful regions according to some criteria. Hierarchical segmentation has emerged as a major trend in this regard as it favors the emergence of important regions at different scales. On the other hand, many methods allow us to have prior information on the position of structures of interest in the images. In this paper, we present a versatile hierarchical segmentation method that takes into account any prior spatial information and outputs a hierarchical segmentation that emphasizes the contours or regions of interest while preserving the important structures in the image. An application of this method to the weakly-supervised segmentation problem is presented.