Weakly Supervised Few-Shot Segmentation Via Meta-Learning

TL;DR

This paper introduces two meta-learning methods, WeaSeL and ProtoSeg, for few-shot semantic segmentation with sparse annotations, demonstrating effectiveness across diverse medical and remote sensing datasets.

Contribution

The paper presents novel meta-learning approaches for few-shot segmentation that work with sparse labels, applicable to medical and remote sensing images.

Findings

Effective segmentation of crops and anatomical parts with sparse annotations

Achieved competitive results compared to fully annotated models

Validated across 12 diverse datasets

Abstract

Semantic segmentation is a classic computer vision task with multiple applications, which includes medical and remote sensing image analysis. Despite recent advances with deep-based approaches, labeling samples (pixels) for training models is laborious and, in some cases, unfeasible. In this paper, we present two novel meta learning methods, named WeaSeL and ProtoSeg, for the few-shot semantic segmentation task with sparse annotations. We conducted extensive evaluation of the proposed methods in different applications (12 datasets) in medical imaging and agricultural remote sensing, which are very distinct fields of knowledge and usually subject to data scarcity. The results demonstrated the potential of our method, achieving suitable results for segmenting both coffee/orange crops and anatomical parts of the human body in comparison with full dense annotation.