Change Detection Between Optical Remote Sensing Imagery and Map Data via Segment Anything Model (SAM)
This addresses the problem of time-sensitive Earth monitoring for remote sensing applications, but it is incremental as it adapts an existing foundation model to a specific task.
The study tackled unsupervised change detection between optical remote sensing imagery and map data by using the Segment Anything Model (SAM) to segment images and compare them in the segmentation domain, achieving competitive results on three datasets compared to existing methods.
Unsupervised multimodal change detection is pivotal for time-sensitive tasks and comprehensive multi-temporal Earth monitoring. In this study, we explore unsupervised multimodal change detection between two key remote sensing data sources: optical high-resolution imagery and OpenStreetMap (OSM) data. Specifically, we propose to utilize the vision foundation model Segmentation Anything Model (SAM), for addressing our task. Leveraging SAM's exceptional zero-shot transfer capability, high-quality segmentation maps of optical images can be obtained. Thus, we can directly compare these two heterogeneous data forms in the so-called segmentation domain. We then introduce two strategies for guiding SAM's segmentation process: the 'no-prompt' and 'box/mask prompt' methods. The two strategies are designed to detect land-cover changes in general scenarios and to identify new land-cover objects within existing backgrounds, respectively. Experimental results on three datasets indicate that the proposed approach can achieve more competitive results compared to representative unsupervised multimodal change detection methods.