Divide, Ensemble and Conquer: The Last Mile on Unsupervised Domain Adaptation for Semantic Segmentation
This addresses the problem of generalizing UDA methods from single-source to multi-source synthetic datasets for semantic segmentation, representing an incremental advancement in the field.
The paper tackles the challenge of unsupervised domain adaptation (UDA) for semantic segmentation across synthetic-to-real domain gaps, proposing the DEC framework that achieves state-of-the-art performance on datasets like Cityscapes, BDD100K, and Mapillary Vistas, significantly narrowing the domain gap.
The last mile of unsupervised domain adaptation (UDA) for semantic segmentation is the challenge of solving the syn-to-real domain gap. Recent UDA methods have progressed significantly, yet they often rely on strategies customized for synthetic single-source datasets (e.g., GTA5), which limits their generalisation to multi-source datasets. Conversely, synthetic multi-source datasets hold promise for advancing the last mile of UDA but remain underutilized in current research. Thus, we propose DEC, a flexible UDA framework for multi-source datasets. Following a divide-and-conquer strategy, DEC simplifies the task by categorizing semantic classes, training models for each category, and fusing their outputs by an ensemble model trained exclusively on synthetic datasets to obtain the final segmentation mask. DEC can integrate with existing UDA methods, achieving state-of-the-art performance on Cityscapes, BDD100K, and Mapillary Vistas, significantly narrowing the syn-to-real domain gap.