Dual-level Interaction for Domain Adaptive Semantic Segmentation

16 Jul 2023  ·  Dongyu Yao, Boheng Li ·

Self-training approach recently secures its position in domain adaptive semantic segmentation, where a model is trained with target domain pseudo-labels. Current advances have mitigated noisy pseudo-labels resulting from the domain gap. However, they still struggle with erroneous pseudo-labels near the boundaries of the semantic classifier. In this paper, we tackle this issue by proposing a dual-level interaction for domain adaptation (DIDA) in semantic segmentation. Explicitly, we encourage the different augmented views of the same pixel to have not only similar class prediction (semantic-level) but also akin similarity relationship with respect to other pixels (instance-level). As it's impossible to keep features of all pixel instances for a dataset, we, therefore, maintain a labeled instance bank with dynamic updating strategies to selectively store the informative features of instances. Further, DIDA performs cross-level interaction with scattering and gathering techniques to regenerate more reliable pseudo-labels. Our method outperforms the state-of-the-art by a notable margin, especially on confusing and long-tailed classes. Code is available at \href{https://github.com/RainJamesY/DIDA}

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Unsupervised Domain Adaptation GTAV-to-Cityscapes Labels DIDA mIoU 71.0 # 6
Unsupervised Domain Adaptation SYNTHIA-to-Cityscapes DIDA mIoU (13 classes) 70.1 # 6
mIoU 63.3 # 5
MIoU (16 classes) 63.3 # 2

Methods


No methods listed for this paper. Add relevant methods here