Cross-Domain Grouping and Alignment for Domain Adaptive Semantic Segmentation
This work provides an incremental improvement in domain adaptive semantic segmentation by better handling multi-modal data distributions and class imbalance, which is important for practitioners deploying models in diverse real-world environments.
This paper addresses the limitation of existing domain adaptation techniques for semantic segmentation that do not account for inter-class variation within the target domain. The authors introduce a learnable clustering module and a novel framework called cross-domain grouping and alignment, which consistently boosts adaptation performance and outperforms state-of-the-art methods across various domain adaptation settings.
Existing techniques to adapt semantic segmentation networks across the source and target domains within deep convolutional neural networks (CNNs) deal with all the samples from the two domains in a global or category-aware manner. They do not consider an inter-class variation within the target domain itself or estimated category, providing the limitation to encode the domains having a multi-modal data distribution. To overcome this limitation, we introduce a learnable clustering module, and a novel domain adaptation framework called cross-domain grouping and alignment. To cluster the samples across domains with an aim to maximize the domain alignment without forgetting precise segmentation ability on the source domain, we present two loss functions, in particular, for encouraging semantic consistency and orthogonality among the clusters. We also present a loss so as to solve a class imbalance problem, which is the other limitation of the previous methods. Our experiments show that our method consistently boosts the adaptation performance in semantic segmentation, outperforming the state-of-the-arts on various domain adaptation settings.