A Sensor Agnostic Domain Generalization Framework for Leveraging Geospatial Foundation Models: Enhancing Semantic Segmentation viaSynergistic Pseudo-Labeling and Generative Learning
This work addresses annotation scarcity and sensor variability in remote sensing for applications like land cover mapping, though it appears incremental as it builds on existing domain adaptation and foundation model approaches.
The paper tackles the problem of limited labeled data and sensor variability in remote sensing semantic segmentation by introducing a domain generalization framework that combines pseudo-labeling with generative pre-training, achieving improved adaptability and segmentation performance as confirmed on hyperspectral and multispectral datasets.
Remote sensing enables a wide range of critical applications such as land cover and land use mapping, crop yield prediction, and environmental monitoring. Advances in satellite technology have expanded remote sensing datasets, yet high-performance segmentation models remain dependent on extensive labeled data, challenged by annotation scarcity and variability across sensors, illumination, and geography. Domain adaptation offers a promising solution to improve model generalization. This paper introduces a domain generalization approach to leveraging emerging geospatial foundation models by combining soft-alignment pseudo-labeling with source-to-target generative pre-training. We further provide new mathematical insights into MAE-based generative learning for domain-invariant feature learning. Experiments with hyperspectral and multispectral remote sensing datasets confirm our method's effectiveness in enhancing adaptability and segmentation.