Confidence-guided Source Label Refinement and Class-aware Pseudo-Label Thresholding for Domain-Adaptive Semantic Segmentation

• Published in: 2024 International Joint Conference on Neural Networks (IJCNN) pp. 1 - 8
• Main Authors: Chu, Hongwei, Shen, Meng, Peng, Jiawen, Ma, Andy J
• Format: Conference Proceeding
• Language: English
• Published: IEEE 30-06-2024
• Subjects: Adaptation models; Adaptive filters; Benchmark testing; Neural networks; Semantic segmentation; Training; Unsupervised Domain Adaptation
• ISSN: 2161-4407
• DOI: 10.1109/IJCNN60899.2024.10650579

Domain-Adaptive Semantic Segmentation (DASS) aims to transfer a segmentation model trained on a labeled source domain to an unlabeled target domain. Most existing methods overlook the errors in the source labels and directly utilize the erroneous source labels for training. This may result in wrong source domain knowledge being wrongly transferred to the target domain, which leads to similar classes cannot be separated well in the target domain. To this end, we propose a novel Confidence-guided Online REfinement (CORE) method, which introduces a refinement network trained on the target data to rectify erroneous source labels online. This can effectively avoid transferring erroneous knowledge from the source domain to the target domain. On the other hand, using the same threshold for all classes in the target domain may result in a severe class imbalance in self-training. Therefore, we propose Class-aware Adaptive Thresholding (CAT) to calculate different thresholds for different target classes, which adaptively filter target pixels with high confidence for better adaptation. Our proposed CORE-CAT method can be easily integrated with existing methods and achieves superior performance across various domain-adaptive semantic segmentation benchmarks.