Brewing Stronger Features: Dual-Teacher Distillation for Multispectral Earth Observation

TL;DR

This paper introduces a dual-teacher contrastive distillation method for multispectral Earth Observation data, enabling effective cross-modal learning and achieving state-of-the-art results across various benchmarks.

Contribution

It proposes a novel dual-teacher contrastive distillation framework that aligns multispectral and optical foundation models for improved cross-modal representation learning.

Findings

Achieves state-of-the-art performance in semantic segmentation, change detection, and classification.

Enables effective multispectral data adaptation without loss on optical-only inputs.

Demonstrates the efficiency of contrastive distillation for heterogeneous Earth Observation data.

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}{https://wolfilip.github.io/DEO/}.