VLM2GeoVec: Toward Universal Multimodal Embeddings for Remote Sensing

Emanuel Sánchez Aimar; Gulnaz Zhambulova; Fahad Shahbaz Khan; Yonghao Xu; Michael Felsberg

Paper

VLM2GeoVec: Toward Universal Multimodal Embeddings for Remote Sensing

Abstract

Satellite imagery differs fundamentally from natural images: its aerial viewpoint, very high resolution, diverse scale variations, and abundance of small objects demand both region-level spatial reasoning and holistic scene understanding. Current remote-sensing approaches remain fragmented between dual-encoder retrieval models, which excel at large-scale cross-modal search but cannot interleave modalities, and generative assistants, which support region-level interpretation but lack scalable retrieval capabilities. We propose

, an instruction-following, single-encoder vision-language model trained contrastively to embed interleaved inputs (images, text, bounding boxes, and geographic coordinates) in a unified vector space. Our single encoder interleaves all inputs into one joint embedding trained with a contrastive loss, eliminating multi-stage pipelines and task-specific modules. To evaluate its versatility, we introduce

, a novel benchmark covering key remote-sensing embedding applications: scene classification; cross-modal search; compositional retrieval; visual-question answering; visual grounding and region-level reasoning; and semantic geospatial retrieval. On RSMEB, it achieves

P@1 on region-caption retrieval (+25 pp vs. dual-encoder baselines),

P@1 on referring-expression retrieval (+19 pp), and

P@1 on semantic geo-localization retrieval (over

prior best), while matching or exceeding specialized baselines on conventional tasks such as scene classification and cross-modal retrieval. VLM2GeoVec unifies scalable retrieval with region-level spatial reasoning, enabling cohesive multimodal analysis in remote sensing. We will publicly release the code, checkpoints, and data upon acceptance.