Diversified Multi-prototype Representation for Semi-supervised Segmentation

TL;DR

This paper introduces a semi-supervised segmentation method that uses multiple prototypes per class and regularization strategies to improve performance with limited labeled data, validated on medical datasets.

Contribution

It proposes a novel multi-prototype representation for classes and regularization techniques to enhance semi-supervised segmentation accuracy.

Findings

Improved segmentation accuracy on medical datasets with few labels

Effective use of mutual information for prototype utilization

Orthogonal prototype enforcement enhances class separation

Abstract

This work considers semi-supervised segmentation as a dense prediction problem based on prototype vector correlation and proposes a simple way to represent each segmentation class with multiple prototypes. To avoid degenerate solutions, two regularization strategies are applied on unlabeled images. The first one leverages mutual information maximization to ensure that all prototype vectors are considered by the network. The second explicitly enforces prototypes to be orthogonal by minimizing their cosine distance. Experimental results on two benchmark medical segmentation datasets reveal our method's effectiveness in improving segmentation performance when few annotated images are available.