Diversified Multi-prototype Representation for Semi-supervised Segmentation
This work addresses the problem of limited labeled data for medical image segmentation, offering an incremental improvement over existing methods.
The paper tackles semi-supervised segmentation by representing each class with multiple prototypes and using regularization strategies to avoid degenerate solutions, achieving improved segmentation performance on medical datasets with few annotated images.
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.