Presentation of the article ‘Boundary-weighted logit consistency improves calibration of segmentation networks’

DOI: 10.48550/arXiv.2307.08163

Abstract of the article:

Neural network prediction probabilities and accuracy are often only weakly-correlated. Inherent label ambiguity in training data for image segmentation aggravates such miscalibration. We show that logit consistency across stochastic transformations acts as a spatially varying regularizer that prevents overconfident predictions at pixels with ambiguous labels. Our boundary-weighted extension of this regularizer provides state-of-the-art calibration for prostate and heart MRI segmentation.