# Inconsistency Masks: Harnessing Model Disagreement for Stable Semi-Supervised Segmentation

arXiv: https://arxiv.org/abs/2401.14387 · HTML: https://mlsanity.com/paper/2401.14387/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 42 references that are in the map.
- First posted on arXiv 25 Jan 2024.

## Builds on (most cited first, 29 of 29)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Improved Regularization of Convolutional Neural Networks with Cutout](https://mlsanity.com/paper/1708.04552/) (2017)
- [Attention U-Net: Learning Where to Look for the Pancreas](https://mlsanity.com/paper/1804.03999/) (2018)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Random Erasing Data Augmentation](https://mlsanity.com/paper/1708.04896/) (2017)
- [Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1902.03368/) (2019)
- [Image Segmentation Using Deep Learning: A Survey](https://mlsanity.com/paper/2001.05566/) (2020)
- [Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations](https://mlsanity.com/paper/1707.03237/) (2017)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning](https://mlsanity.com/paper/1908.02983/) (2019)
- [Snapshot Ensembles: Train 1, get M for free](https://mlsanity.com/paper/1704.00109/) (2017)
- [In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning](https://mlsanity.com/paper/2101.06329/) (2021)
- [Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation](https://mlsanity.com/paper/1908.10454/) (2019)
- [Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels](https://mlsanity.com/paper/2203.03884/) (2022)
- [ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentation](https://mlsanity.com/paper/2106.05095/) (2021)
- [Revisiting Weak-to-Strong Consistency in Semi-Supervised Semantic Segmentation](https://mlsanity.com/paper/2208.09910/) (2022)
- [Semantic Segmentation of Underwater Imagery: Dataset and Benchmark](https://mlsanity.com/paper/2004.01241/) (2020)
- [Instance-specific and Model-adaptive Supervision for Semi-supervised Semantic Segmentation](https://mlsanity.com/paper/2211.11335/) (2022)
- [Fuzzy Positive Learning for Semi-supervised Semantic Segmentation](https://mlsanity.com/paper/2210.08519/) (2022)

## Abstract (arXiv)

A primary challenge in semi-supervised learning (SSL) for segmentation is the confirmation bias from noisy pseudo-labels, which destabilizes training and degrades performance. We propose Inconsistency Masks (IM), a framework that reframes model disagreement not as noise to be averaged away, but as a valuable signal for identifying uncertainty. IM leverages an ensemble of teacher models to generate a mask that explicitly delineates regions where predictions diverge. By filtering these inconsistent areas from input-pseudo-label pairs, our method effectively mitigates the cycle of error propagation common in both continuous and iterative self-training paradigms. Extensive experiments on the Cityscapes benchmark demonstrate IM's effectiveness as a general enhancement framework: when paired with leading approaches like iMAS, U2PL, and UniMatch, our method consistently boosts accuracy, achieving superior benchmarks across ResNet-50 and DINOv2 backbones, and even improving distilled architectures like SegKC. Furthermore, the method's robustness is confirmed in resource-constrained scenarios where pre-trained weights are unavailable. On three additional diverse datasets from medical and underwater domains trained entirely from scratch, IM significantly outperforms standard SSL baselines. Notably, the IM framework is dataset-agnostic, seamlessly handling binary, multi-class, and complex multi-label tasks by operating on discretized predictions. By prioritizing training stability, IM offers a generalizable and robust solution for semi-supervised segmentation, particularly in specialized areas lacking large-scale pre-training data. The full code is available at: https://github.com/MichaelVorndran/InconsistencyMasks