Hot-Distance: Combining One-Hot and Signed Distance Embeddings for Segmentation

TL;DR

This paper introduces Hot-Distance, a new segmentation target that combines one-hot encoding with signed boundary distance prediction to enhance training data usability for subcellular structure segmentation in FIB-SEM images.

Contribution

It proposes a novel segmentation target that merges boundary distance and one-hot encoding, improving data utilization in microscopy image segmentation.

Findings

Increased training data usability for segmentation tasks.

Improved segmentation accuracy in FIB-SEM images.

Enhanced model robustness with the new target.

Abstract

Machine learning models are only as good as the data to which they are fit. As such, it is always preferable to use as much data as possible in training models. What data can be used for fitting a model depends a lot on the formulation of the task. We introduce Hot-Distance, a novel segmentation target that incorporates the strength of signed boundary distance prediction with the flexibility of one-hot encoding, to increase the amount of usable training data for segmentation of subcellular structures in focused ion beam scanning electron microscopy (FIB-SEM).