A Sensor Agnostic Domain Generalization Framework for Leveraging Geospatial Foundation Models: Enhancing Semantic Segmentation viaSynergistic Pseudo-Labeling and Generative Learning

TL;DR

This paper proposes a domain generalization framework for remote sensing semantic segmentation that combines pseudo-labeling and generative learning to improve model adaptability across diverse sensors and conditions.

Contribution

It introduces a novel approach integrating pseudo-labeling with generative pre-training for geospatial foundation models, supported by new mathematical insights into MAE-based learning.

Findings

Enhanced segmentation accuracy across different sensors and environments.

Effective domain adaptation demonstrated on hyperspectral and multispectral datasets.

Mathematical analysis clarifies the role of MAE in domain-invariant feature learning.

Abstract

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.