Abstract:By reason of factors such as terrains, weather conditions, sensor imaging methods, and cultural and economic development, there is a large shift between the remote sensing imagery collected from different geographic locations and different sensors, which makes the state-of-the-art semantic segmentation models trained on source domain (an image set gathered from specific geographic locations and sensors) difficult to generalize to target domain (another image set collected from other geographic locations and sensors). Currently, unsupervised domain adaptation (UDA) using adversarial training, whose purpose is to align the marginal distribution in the output space between the source and target domains, is the most explored and practical approach to address this issue. However, this global alignment approach does not take into account diversities of different regions in a specific image nor the category-level distribution, which leads to the consequence that some regions and categories which are already well aligned between the source and target domains may be incorrectly remapped. Therefore, we propose a region and category adaptive domain discriminator (RCA-DD), aiming to emphasize the differences in regions and categories during the process of alignment. Specifically, on the one hand, we propose an entropy-based regional attention module (ERAM) in domain discriminator to emphasize the importance of difficult-to-align regions. On the other hand, we propose a class-clear module (CCM) to update only the distribution of existing categories in one iteration without affecting all categories. Finally, a lot of experiments are introduced to indicate that the proposed method can obtain better results when compared with other state-of-the-art UDA methods using adversarial training.

Semantics-Preserved Domain Adaptation with Target Diverse Perturbation and Test Ensembling for Image Segmentation

Weakly-Supervised Domain Adaptation With Adversarial Entropy for Building Segmentation in Cross-Domain Aerial Imagery

Unsupervised Domain Adaptive Image Semantic Segmentation Based on Convolutional Fine-Grained Discriminant and Entropy Minimization

Unsupervised Domain Adaptation for Referring Semantic Segmentation

Partial Domain Adaptation on Semantic Segmentation

Domain Adaptation for Remote Sensing Image Semantic Segmentation: An Integrated Approach of Contrastive Learning and Adversarial Learning

March on Data Imperfections: Domain Division and Domain Generalization for Semantic Segmentation.

Uncertainty-Aware Consistency Regularization for Cross-Domain Semantic Segmentation

Threshold-adaptive Unsupervised Focal Loss for Domain Adaptation of Semantic Segmentation

Feature Re-Representation and Reliable Pseudo Label Retraining for Cross-Domain Semantic Segmentation.

Domain Adaptation for Semantic Segmentation of Road Scenes Via Two-Stage Alignment of Traffic Elements

Semi-supervised Domain Adaptation for Semantic Segmentation

Unsupervised Domain Adaptation for Remote Sensing Image Semantic Segmentation Using Region and Category Adaptive Domain Discriminator

Differential Treatment for Stuff and Things: A Simple Unsupervised Domain Adaptation Method for Semantic Segmentation

A Fine-Grained Unsupervised Domain Adaptation Framework for Semantic Segmentation of Remote Sensing Images

Unsupervised cross domain semantic segmentation with mutual refinement and information distillation

Entropy Guided Adversarial Domain Adaptation for Aerial Image Semantic Segmentation

Cross-Domain Scene Unsupervised Learning Segmentation with Dynamic Subdomains

Toward few-shot domain adaptation with perturbation-invariant representation and transferable prototypes

Enhancing Semi-Supervised Semantic Segmentation of Remote Sensing Images via Feature Perturbation-Based Consistency Regularization Methods

Unsupervised Domain Adaptation for Semantic Segmentation of High-resolution Remote Sensing Imagery Driven by Category-Certainty Attention