mlsanityData to 14 Oct 2026

Spatially Varying Label Smoothing: Capturing Uncertainty from Expert Annotations

arXiv 2104.05788 ↗First posted 12 Apr 2021cs.CVcited by 31

Not on the map. Why?

Abstract (arXiv)

The task of image segmentation is inherently noisy due to ambiguities regarding the exact location of boundaries between anatomical structures. We argue that this information can be extracted from the expert annotations at no extra cost, and when integrated into state-of-the-art neural networks, it can lead to improved calibration between soft probabilistic predictions and the underlying uncertainty. We built upon label smoothing (LS) where a network is trained on 'blurred' versions of the ground truth labels which has been shown to be effective for calibrating output predictions. However, LS is not taking the local structure into account and results in overly smoothed predictions with low confidence even for non-ambiguous regions. Here, we propose Spatially Varying Label Smoothing (SVLS), a soft labeling technique that captures the structural uncertainty in semantic segmentation. SVLS also naturally lends itself to incorporate inter-rater uncertainty when multiple labelmaps are available. The proposed approach is extensively validated on four clinical segmentation tasks with different imaging modalities, number of classes and single and multi-rater expert annotations. The results demonstrate that SVLS, despite its simplicity, obtains superior boundary prediction with improved uncertainty and model calibration.

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). Builds on 22 of its 36 references that are in the map. First posted on arXiv 12 Apr 2021.

Builds on

22 of 36 references.

References · 22

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  4. 2017On Calibration of Modern Neural Networkscited by 4,960
  5. 20163D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotationcited by 1,476
  6. 2019When Does Label Smoothing Help?cited by 988
  7. 2018Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challengecited by 511
  8. 2017Regularizing Neural Networks by Penalizing Confident Output Distributionscited by 464
  9. 2019Measuring Calibration in Deep Learningcited by 446
  10. 2019Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibrationcited by 364
  11. 2019The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomescited by 172
  12. 2018Exploring Uncertainty Measures in Deep Networks for Multiple Sclerosis Lesion Detection and Segmentationcited by 146
10 more
  1. 2019PHiSeg: Capturing Uncertainty in Medical Image Segmentationcited by 128
  2. 2020Stochastic Segmentation Networks: Modelling Spatially Correlated Aleatoric Uncertaintycited by 101
  3. 2016Towards better decoding and language model integration in sequence to sequence modelscited by 100
  4. 2018Deep learning to achieve clinically applicable segmentation of head and neck anatomy for radiotherapycited by 77
  5. 2019Non-Parametric Calibration for Classificationcited by 57
  6. 2014Multi-Atlas Segmentation of Biomedical Images: A Surveycited by 56
  7. 2018On the Effect of Inter-observer Variability for a Reliable Estimation of Uncertainty of Medical Image Segmentationcited by 46
  8. 2019Well-calibrated Model Uncertainty with Temperature Scaling for Dropout Variational Inferencecited by 36
  9. 2019Uncertainty Measures and Prediction Quality Rating for the Semantic Segmentation of Nested Multi Resolution Street Scene Imagescited by 18
  10. 2019TuNet: End-to-end Hierarchical Brain Tumor Segmentation using Cascaded Networkscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 3; 2022: 6; 2023: 7; 2024: 6; 2025: 4; 2026: 5.

31 papers.

Sort:

By area:

  1. 2023A review of uncertainty quantification in medical image analysis: probabilistic and non-probabilistic methodscited by 39
  2. 2024Tyche: Stochastic In-Context Learning for Medical Image Segmentationcited by 20
  3. 2021Label Cleaning Multiple Instance Learning: Refining Coarse Annotations on Single Whole-Slide Imagescited by 13
  4. 2021Maximum Entropy on Erroneous Predictions (MEEP): Improving model calibration for medical image segmentationcited by 13
  5. 2023Trust your neighbours: Penalty-based constraints for model calibrationcited by 12
  6. 2022On the Optimal Combination of Cross-Entropy and Soft Dice Losses for Lesion Segmentation with Out-of-Distribution Robustnesscited by 11
  7. 2023Multi-Head Multi-Loss Model Calibrationcited by 8
  8. 2022Learning to segment fetal brain tissue from noisy annotationscited by 7
  9. 2024Neighbor-Aware Calibration of Segmentation Networks with Penalty-Based Constraintscited by 7
  10. 2025Calibration and Uncertainty for multiRater Volume Assessment in multiorgan Segmentation (CURVAS) challenge resultscited by 7
  11. 2023Paced-Curriculum Distillation with Prediction and Label Uncertainty for Image Segmentationcited by 6
  12. 2023Dice Semimetric Losses: Optimizing the Dice Score with Soft Labelscited by 5
19 more
  1. 2021Deep Quantile Regression for Uncertainty Estimation in Unsupervised and Supervised Lesion Detectioncited by 4
  2. 2023Boundary-weighted logit consistency improves calibration of segmentation networkscited by 4
  3. 2022Weakly Supervised Medical Image Segmentation With Soft Labels and Noise Robust Losscited by 2
  4. 2024Class and Region-Adaptive Constraints for Network Calibrationcited by 2
  5. 2025We Care Each Pixel: Calibrating on Medical Segmentation Modelcited by 2
  6. 2022Context Label Learning: Improving Background Class Representations in Semantic Segmentationcited by 1
  7. 2022Confidence-Aware Paced-Curriculum Learning by Label Smoothing for Surgical Scene Understandingcited by 1
  8. 2024Advancing Medical Image Segmentation: Morphology-Driven Learning with Diffusion Transformercited by 1
  9. 2024Improving Uncertainty-Error Correspondence in Deep Bayesian Medical Image Segmentationcited by 1
  10. 2025A Lightweight and Extensible Cell Segmentation and Classification Model for Whole Slide Imagescited by 1
  11. 2026SNCE: Geometry-Aware Supervision for Scalable Discrete Image Generationcited by 1
  12. 2026Beyond Boundary Noise: Aggregated Aleatoric Uncertainty Fails to Capture Presence Ambiguity in 3D Lung Nodule Segmentationcited by 1
  13. 2022Learning with Weak Annotations for Robust Maritime Obstacle Detectioncited by 0
  14. 2023Can ultrasound confidence maps predict sonographers' labeling variability?cited by 0
  15. 2024OBSeg: Accurate and Fast Instance Segmentation Framework Using Segmentation Foundation Models with Oriented Bounding Box Promptscited by 0
  16. 2025Good Enough? An Investigation on the Impact of Label Quality in Large-Scale Medical Datasetscited by 0
  17. 2026Assessing Pancreatic Ductal Adenocarcinoma Vascular Invasion: the PDACVI Benchmarkcited by 0
  18. 2026Multi-Rater Calibrated Segmentation Modelscited by 0
  19. 2026Beyond scalar losses: calibrating segmentation models via gradient vector field surgerycited by 0