cs.CVNov 10, 2026

Satellite Pretraining Is Not One Thing: A Matched-Pair Study of Generic and Earth-Observation Foundation Models for Scene Classification

Authors: Sandarbh Agrawal, Rakshith Sathish, Ashwathi Nambiar

Organizations: SatSure Analytics India Pvt. Ltd., Bengaluru, India

Abstract

Geospatial foundation models are pretrained on satellite and aerial imagery, but whether such an encoder beats a strong generic one for few-label land-use/land-cover mapping remains unclear, as prior comparisons rarely isolate the pretraining data. We study five matched pairs of Earth-observation (EO) and generic encoders sharing objective and architecture (DINO/SatDINO, DINOv3/DINOv3-SAT, MAE/SatMAE, MAE/Scale-MAE, CLIP/RemoteCLIP) within a benchmark of 14 frozen encoders on five RGB scene-classification datasets (0.06-30 m GSD). With native preprocessing, we evaluate k-NN, linear and attentive probes and image retrieval over six low-label splits, using paired per-image significance tests. EO pretraining is not uniformly helpful: on most datasets and metrics, it significantly improves three pairs (SatDINO, Scale-MAE, RemoteCLIP, the latter evaluated only on unseen datasets) and significantly hurts two (linear probe: DINOv3-SAT by up to 22.5 points, SatMAE by up to 8.8), with opposite outcomes even for two MAEs pretrained on the same fMoW-RGB imagery. These effects are robust to patch-token pooling and attentive probing, concentrate on small man-made objects, and are largest with 1-5% labels. Across encoders, the silhouette score of frozen embeddings predicts k-NN accuracy and retrieval mAP almost perfectly (Spearman ρ≥0.96ρ\geq 0.96 on four of five datasets). Pretraining on satellite imagery is thus not a reliable selection criterion; candidate checkpoints must be validated on target data.

Figures & tables

Explore similar work

CardsList
  1. A Composition-Aware Pretraining Framework for Geospatial Foundation Models

    Aug 31, 2026Aryan Kashyap Naveen, Abhishek Srinivas, Pranav Moothedath +1Remote Sensing Image UnderstandingRepresentation Learning

  2. When GeoFMs Meet VHR: Benchmarking Foundation Models on Very High Resolution Earth Observation Imagery

    Nov 5, 2026Kshitij Raj Sharma, Pierre Adorni, Minh-Tan Pham +2Geospatial Foundation ModelsRemote Sensing

  3. Do Satellites See Commuters? A Critical Benchmark of Vision Foundation Models

    Sep 1, 2026Ashiq Shukoor Iqbal, Wilson Wongso, Flora D. SalimVision Foundation ModelsGeospatial Foundation Models