---
license: cc-by-4.0
tags:
  - earth-observation
  - remote-sensing
  - sentinel-2
  - sentinel-1
  - embeddings
  - model-distillation
  - geospatial
datasets:
  - Major-TOM/Core-S2-L1C
  - Major-TOM/Core-S2-L2A
  - Major-TOM/Core-S1-RTC
  - Major-TOM/Core-AlphaEarth-Embeddings
library_name: betaearth
pipeline_tag: feature-extraction
---

# betaearth-segformer

BetaEarth SegFormer-B2 no FiLM (ISPRS baseline) — no timestamp needed

Part of the **BetaEarth** family — fully trainable, without temporal conditioning.

| Metric | Value |
|--------|-------|
| Test cosine similarity | 0.88 |
| LULC downstream accuracy | 0.869 |
| Trainable parameters | 104.8M |
| Total parameters | 104.8M |
| Inputs | S2 L1C+L2A (9ch), S1 RTC (2ch), COP-DEM (1ch) |
| Output | (H, W, 64) float32, L2-normalised |

## Usage

```bash
pip install betaearth
```

```python
from betaearth import BetaEarth

model = BetaEarth.from_pretrained("asterisk-labs/betaearth-segformer")
embedding = model.predict(
    s2_l2a=s2_l2a,   # (9, H, W) uint16
    s1=s1,            # (2, H, W) float32
    dem=dem,          # (1, H, W) float32
    doy=182,
)
# embedding: (H, W, 64) numpy array
```

## All BetaEarth models

| Model | Cos Sim | Params | Best for |
|-------|---------|--------|----------|
| [betaearth-segformer-film](asterisk-labs/betaearth-segformer-film) | **0.886** | 0.3M | Best quality |
| [betaearth-segformer-film-hilr](asterisk-labs/betaearth-segformer-film-hilr) | 0.886 | 0.3M | Alt frozen |
| [betaearth-segformer](asterisk-labs/betaearth-segformer) | 0.880 | 104.8M | No timestamp |
| [betaearth-segformer-film-scratch](asterisk-labs/betaearth-segformer-film-scratch) | 0.883 | 104.8M | End-to-end |
| [betaearth-rgb-only](asterisk-labs/betaearth-rgb-only) | 0.836 | 26.3M | Minimal data |

## Citation

```bibtex
@inproceedings{czerkawski2026betaearth,
  title     = {BetaEarth: Emulating Closed-Source Earth Observation Foundation Models Through Their Public Embeddings},
  author    = {Czerkawski, Mikolaj},
  booktitle = {ISPRS Congress 2026},
  year      = {2026}
}
```

## License

CC-BY 4.0. Training data attribution: "The AlphaEarth Foundations Satellite Embedding dataset is produced by Google and Google DeepMind."