Decomposition-based Unsupervised Domain Adaptation for Remote Sensing Image Semantic Segmentation

TL;DR

This paper introduces a novel unsupervised domain adaptation framework for remote sensing image segmentation that decomposes features into frequency components and uses a global-local GAN with transformers to improve cross-domain transferability.

Contribution

It proposes a multiscale high/low-frequency decomposition scheme integrated into a global-local GAN with transformers for enhanced domain adaptation in remote sensing segmentation.

Findings

Outperforms existing UDA methods on benchmark datasets.

Effectively preserves local details and global semantics.

Demonstrates improved cross-domain transferability.

Abstract

Unsupervised domain adaptation (UDA) techniques are vital for semantic segmentation in geosciences, effectively utilizing remote sensing imagery across diverse domains. However, most existing UDA methods, which focus on domain alignment at the high-level feature space, struggle to simultaneously retain local spatial details and global contextual semantics. To overcome these challenges, a novel decomposition scheme is proposed to guide domain-invariant representation learning. Specifically, multiscale high/low-frequency decomposition (HLFD) modules are proposed to decompose feature maps into high- and low-frequency components across different subspaces. This decomposition is integrated into a fully global-local generative adversarial network (GLGAN) that incorporates global-local transformer blocks (GLTBs) to enhance the alignment of decomposed features. By integrating the HLFD scheme and the GLGAN, a novel decomposition-based UDA framework called De-GLGAN is developed to improve the cross-domain transferability and generalization capability of semantic segmentation models. Extensive experiments on two UDA benchmarks, namely ISPRS Potsdam and Vaihingen, and LoveDA Rural and Urban, demonstrate the effectiveness and superiority of the proposed approach over existing state-of-the-art UDA methods. The source code for this work is accessible at https://github.com/sstary/SSRS.