Class Balanced Dynamic Acquisition for Domain Adaptive Semantic Segmentation using Active Learning
This addresses label efficiency and class imbalance in semantic segmentation for computer vision applications, representing an incremental improvement over existing methods.
The paper tackles the class imbalance issue in domain adaptive active learning for semantic segmentation, introducing Class Balanced Dynamic Acquisition (CBDA) which improves performance, achieving gains of 0.6 to 2.4 mIoU over baselines for different budgets and boosting minority class IoU by up to 4.6.
Domain adaptive active learning is leading the charge in label-efficient training of neural networks. For semantic segmentation, state-of-the-art models jointly use two criteria of uncertainty and diversity to select training labels, combined with a pixel-wise acquisition strategy. However, we show that such methods currently suffer from a class imbalance issue which degrades their performance for larger active learning budgets. We then introduce Class Balanced Dynamic Acquisition (CBDA), a novel active learning method that mitigates this issue, especially in high-budget regimes. The more balanced labels increase minority class performance, which in turn allows the model to outperform the previous baseline by 0.6, 1.7, and 2.4 mIoU for budgets of 5%, 10%, and 20%, respectively. Additionally, the focus on minority classes leads to improvements of the minimum class performance of 0.5, 2.9, and 4.6 IoU respectively. The top-performing model even exceeds the fully supervised baseline, showing that a more balanced label than the entire ground truth can be beneficial.