LabOR: Labeling Only if Required for Domain Adaptive Semantic Segmentation
This work addresses the challenge of high annotation costs in domain adaptation for semantic segmentation, offering a practical solution to minimize human labor while maintaining performance, though it is incremental in improving existing UDA approaches.
The paper tackles the problem of reducing the performance gap between unsupervised domain adaptation and fully supervised methods in semantic segmentation by introducing a human-in-the-loop strategy that labels only uncertain points, achieving near supervised performance with only about 2.2% of ground truth labels or as few as 40 points.
Unsupervised Domain Adaptation (UDA) for semantic segmentation has been actively studied to mitigate the domain gap between label-rich source data and unlabeled target data. Despite these efforts, UDA still has a long way to go to reach the fully supervised performance. To this end, we propose a Labeling Only if Required strategy, LabOR, where we introduce a human-in-the-loop approach to adaptively give scarce labels to points that a UDA model is uncertain about. In order to find the uncertain points, we generate an inconsistency mask using the proposed adaptive pixel selector and we label these segment-based regions to achieve near supervised performance with only a small fraction (about 2.2%) ground truth points, which we call "Segment based Pixel-Labeling (SPL)". To further reduce the efforts of the human annotator, we also propose "Point-based Pixel-Labeling (PPL)", which finds the most representative points for labeling within the generated inconsistency mask. This reduces efforts from 2.2% segment label to 40 points label while minimizing performance degradation. Through extensive experimentation, we show the advantages of this new framework for domain adaptive semantic segmentation while minimizing human labor costs.