Inconsistency Masks: Removing the Uncertainty from Input-Pseudo-Label Pairs
This addresses the problem of high labeling costs in image segmentation for researchers and practitioners in resource-constrained environments, offering an incremental improvement over existing semi-supervised methods.
The study tackled the bottleneck of limited labeled data for image segmentation by introducing Inconsistency Masks to filter uncertainty in pseudo-label pairs, achieving strong results with only 10% labeled data and even outperforming fully labeled models on the ISIC 2018 dataset.
Efficiently generating sufficient labeled data remains a major bottleneck in deep learning, particularly for image segmentation tasks where labeling requires significant time and effort. This study tackles this issue in a resource-constrained environment, devoid of extensive datasets or pre-existing models. We introduce Inconsistency Masks (IM), a novel approach that filters uncertainty in image-pseudo-label pairs to substantially enhance segmentation quality, surpassing traditional semi-supervised learning techniques. Employing IM, we achieve strong segmentation results with as little as 10% labeled data, across four diverse datasets and it further benefits from integration with other techniques, indicating broad applicability. Notably on the ISIC 2018 dataset, three of our hybrid approaches even outperform models trained on the fully labeled dataset. We also present a detailed comparative analysis of prevalent semi-supervised learning strategies, all under uniform starting conditions, to underline our approach's effectiveness and robustness. The full code is available at: https://github.com/MichaelVorndran/InconsistencyMasks