GeoSAM-Lite: A Lightweight Foundation Model for Onboard Remote Sensing Segmentation

2026-07-04

AI Summary

This work addresses the challenges of deploying large foundation models on resource-constrained remote sensing platforms, particularly their high computational cost and the domain shift between natural and remote sensing imagery. To overcome these issues, we propose a lightweight, prompt-free framework for remote sensing image segmentation. Our approach introduces a geospatial domain initialization (Geo-Init) strategy to bridge the domain gap and incorporates a feature fusion layer (FFL) to recover high-frequency boundary details. Combined with a compact backbone network and domain-specific knowledge distillation, the method enables efficient, real-time segmentation on edge devices. Evaluated on mainstream remote sensing benchmarks, it achieves segmentation accuracy comparable to heavyweight models while reducing parameter count by 92.8%, substantially advancing the Pareto frontier between efficiency and fidelity.

Abstract

The deployment of large-scale foundation models like Segment Anything Model (SAM) on resource-constrained Earth observation platforms is hindered by prohibitive computational costs and the domain shift between natural and remote sensing imagery. To address these challenges, we propose \textit{Geo}spatial \textit{S}egment \textit{A}nything \textit{M}odel-Lite (GeoSAM-Lite), a lightweight, prompt-free segmentation framework designed for efficient onboard remote sensing segmentation. GeoSAM-Lite incorporates two core innovations: (1) Geospatial-Domain Initialization (Geo-Init), a domain-aware pre-training strategy that distills geospatial priors from a specialized teacher to bridge the domain gap; and (2) Feature Fusion Layers (FFL), which recalibrate spatial features and restore high-frequency boundary cues to overcome the capacity bottlenecks of lightweight backbones. Experiments across representative datasets, with a primary focus on cloud scenarios to evaluate performance under extreme scale variations and complex boundaries, demonstrate that GeoSAM-Lite achieves competitive accuracy while reducing parameters by 92.8\% compared to the heavyweight RSAM-Seg. By establishing a superior Pareto frontier between efficiency and fidelity, GeoSAM-Lite offers a practical solution for real-time segmentation on edge devices.

Problem

• foundation model
• remote sensing segmentation
• domain shift
• computational cost
• onboard deployment

Innovation

• lightweight foundation model
• geospatial-domain initialization
• feature fusion layers
• onboard remote sensing segmentation
• domain shift