Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2509.10919 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914036220690432 |
|---|---|
| author | Albughdadi, Mohanad |
| author_facet | Albughdadi, Mohanad |
| contents | Recent advances in Earth Observation have focused on large-scale foundation models. However, these models are computationally expensive, limiting their accessibility and reuse for downstream tasks. In this work, we investigate compact architectures as a practical pathway toward smaller general-purpose EO models. We propose a Metadata-aware Mixture-of-Experts Masked Autoencoder (MoE-MAE) with only 2.5M parameters. The model combines sparse expert routing with geo-temporal conditioning, incorporating imagery alongside latitude/longitude and seasonal/daily cyclic encodings. We pretrain the MoE-MAE on the BigEarthNet-Landsat dataset and evaluate embeddings from its frozen encoder using linear probes. Despite its small size, the model competes with much larger architectures, demonstrating that metadata-aware pretraining improves transfer and label efficiency. To further assess generalization, we evaluate on the EuroSAT-Landsat dataset, which lacks explicit metadata, and still observe competitive performance compared to models with hundreds of millions of parameters. These results suggest that compact, metadata-aware MoE-MAEs are an efficient and scalable step toward future EO foundation models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_10919 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lightweight Metadata-Aware Mixture-of-Experts Masked Autoencoder for Earth Observation Albughdadi, Mohanad Computer Vision and Pattern Recognition Machine Learning Recent advances in Earth Observation have focused on large-scale foundation models. However, these models are computationally expensive, limiting their accessibility and reuse for downstream tasks. In this work, we investigate compact architectures as a practical pathway toward smaller general-purpose EO models. We propose a Metadata-aware Mixture-of-Experts Masked Autoencoder (MoE-MAE) with only 2.5M parameters. The model combines sparse expert routing with geo-temporal conditioning, incorporating imagery alongside latitude/longitude and seasonal/daily cyclic encodings. We pretrain the MoE-MAE on the BigEarthNet-Landsat dataset and evaluate embeddings from its frozen encoder using linear probes. Despite its small size, the model competes with much larger architectures, demonstrating that metadata-aware pretraining improves transfer and label efficiency. To further assess generalization, we evaluate on the EuroSAT-Landsat dataset, which lacks explicit metadata, and still observe competitive performance compared to models with hundreds of millions of parameters. These results suggest that compact, metadata-aware MoE-MAEs are an efficient and scalable step toward future EO foundation models. |
| title | Lightweight Metadata-Aware Mixture-of-Experts Masked Autoencoder for Earth Observation |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2509.10919 |