Exploring Foundation Models in Remote Sensing Image Change Detection: A Comprehensive Survey

TL;DR

This survey addresses the integration of foundation models into remote sensing change detection, outlining the problem, key methods, and practical implications. It offers a two-axis taxonomy by data modality and network architecture, and reviews representative datasets and state-of-the-art foundation-model–driven approaches. The paper highlights the dominance of encoder-decoder and multimodal architectures, the promise of vision-language and heterogeneous models, and the critical challenges of data scarcity, domain shift, and interpretability. It further identifies future directions in data generation, cross-domain adaptation, explainability, and advanced multimodal fusion to boost real-world applicability.

Abstract

Change detection, as an important and widely applied technique in the field of remote sensing, aims to analyze changes in surface areas over time and has broad applications in areas such as environmental monitoring, urban development, and land use analysis.In recent years, deep learning, especially the development of foundation models, has provided more powerful solutions for feature extraction and data fusion, effectively addressing these complexities. This paper systematically reviews the latest advancements in the field of change detection, with a focus on the application of foundation models in remote sensing tasks.