mlsanityData to 14 Oct 2026

Boundary-weighted logit consistency improves calibration of segmentation networks

arXiv 2307.08163 ↗First posted 16 Jul 2023cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 23 of its 38 references that are in the map. First posted on arXiv 16 Jul 2023.

Builds on

23 of 38 references.

References · 23

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2017On Calibration of Modern Neural Networkscited by 4,960
  3. 2019When Does Label Smoothing Help?cited by 988
  4. 2019Measuring Calibration in Deep Learningcited by 446
  5. 2016Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learningcited by 441
  6. 2018A Probabilistic U-Net for Segmentation of Ambiguous Imagescited by 283
  7. 2018Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networkscited by 280
  8. 2019Confidence Calibration and Predictive Uncertainty Estimation for Deep Medical Image Segmentationcited by 163
  9. 2021Ensemble of Averages: Improving Model Selection and Boosting Performance in Domain Generalizationcited by 144
  10. 2019Transformation Consistent Self-ensembling Model for Semi-supervised Medical Image Segmentationcited by 140
  11. 2019PHiSeg: Capturing Uncertainty in Medical Image Segmentationcited by 128
  12. 2022SynthStrip: Skull-Stripping for Any Brain Imagecited by 127
11 more
  1. 2021Mutual Consistency Learning for Semi-supervised Medical Image Segmentationcited by 108
  2. 2020Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertaintycited by 101
  3. 2021The Devil is in the Margin: Margin-based Label Smoothing for Network Calibrationcited by 91
  4. 2021Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labelscited by 74
  5. 2019Semi-Supervised Medical Image Segmentation via Learning Consistency under Transformationscited by 71
  6. 2019Semi-Supervised Brain Lesion Segmentation with an Adapted Mean Teacher Modelcited by 53
  7. 2021Spatially Varying Label Smoothing: Capturing Uncertainty from Expert Annotationscited by 31
  8. 2021Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentationcited by 13
  9. 2023Trust your neighbours: Penalty-based constraints for model calibrationcited by 12
  10. 2022Automatic Segmentation of the Placenta in BOLD MRI Time Seriescited by 5
  11. 2021Hypernet-Ensemble Learning of Segmentation Probability for Medical Image Segmentation with Ambiguous Labelscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 1; 2026: 1.

4 papers.

  1. 2024Class and Region-Adaptive Constraints for Network Calibrationcited by 2
  2. 2023Consistency Regularization Improves Placenta Segmentation in Fetal EPI MRI Time Seriescited by 1
  3. 2025Explaining Uncertainty in Multiple Sclerosis Cortical Lesion Segmentation Beyond Prediction Errorscited by 0
  4. 2026Beyond scalar losses: calibrating segmentation models via gradient vector field surgerycited by 0