[Submitted on 23 Feb 2026 (v1), last revised 21 Jul 2026 (this version, v3)]

View PDF HTML (experimental)

Abstract:Foundation models are transforming Earth Observation (EO), yet the diversity of EO sensors and modalities makes a single universal model unrealistic. Multiple specialized EO foundation models (EOFMs) will likely coexist, making efficient knowledge transfer across modalities essential. Most existing EO pretraining relies on masked image modeling, which emphasizes local reconstruction but provides limited control over global semantic structure. To address this, we propose a dual-teacher contrastive distillation framework for multispectral imagery that aligns the student's pretraining objective with the contrastive self-distillation paradigm of modern optical vision foundation models (VFMs). Our approach combines a multispectral teacher with an optical VFM teacher, enabling coherent cross-modal representation learning. Experiments across diverse optical and multispectral benchmarks show that our model adapts to multispectral data without compromising performance on optical-only inputs, achieving state-of-the-art results in both settings, with an average improvement of 3.64 percentage points in semantic segmentation, 1.2 in change detection, and 1.31 in classification tasks. This demonstrates that contrastive distillation provides a principled and efficient approach to scalable representation learning across heterogeneous EO data sources. Project page: \textcolor{magenta}{this https URL}.

Submission history

From: Filip Wolf [view email]
[v1] Mon, 23 Feb 2026 14:09:01 UTC (7,229 KB)
[v2] Tue, 24 Feb 2026 11:57:03 UTC (7,229 KB)
[v3] Tue, 21 Jul 2026 07:37:55 UTC (7,232 KB)