arjunrao2000/geolayers

Geolayers-Data

This dataset card contains usage instructions and metadata for all data-products released with our paper:
Using Multiple Input Modalities can Improve Data-Efficiency and O.O.D. Generalization for ML with Satellite Imagery. We release 3 modified versions of 3 benchmark datasets spanning land-cover segmentation, tree-cover regression, and multi-label land-cover classification tasks. These datasets are augmented with auxiliary, geographic inputs. A full list of contributed data products is shown in the table below.

Dataset	Task Description	Multispectral Input	Model	Additional Data Layers	OOD Test Set Present?
SustainBench	Farmland boundary delineation	Sentinel-2 RGB	U-Net	OSM rasters, EU-DEM	✗
EnviroAtlas	Land-cover segmentation	NAIP RGB + NIR	FCN	Prior, OSM rasters	✓
BigEarthNet v2.0	Land-cover classification	Sentinel-2 (10 bands)	ViT	SatCLIP embeddings	✓
USAVars	Tree-cover regression	NAIP RGB + NIR	ResNet-50	OSM rasters, DEM	✗

📦 Datasets & Georeferenced Auxiliary Layers

SustainBench – Farmland Boundary Delineation

• Optical input: Sentinel-2 RGB patches (224×224 px, 10 m GSD) covering French cropland in 2017; ≈ 1.6 k training images.
• Auxiliary layers (all geo-aligned):
  • 19-channel OpenStreetMap (OSM) raster stack (roads, waterways, buildings, biome classes, …)
  • EU-DEM (20 m GSD, down-sampled to 10 m)
• Why: OSM + DEM give an 8 % Dice boost when labels are scarce; gains appear once the training set drops below ≈ 700 images.

---

EnviroAtlas – Land-Cover Segmentation

• Optical input: NAIP 4-band RGB-NIR aerial imagery at 1 m resolution.
• Auxiliary layers:
  • OSM rasters (roads, waterbodies, waterways)
  • Prior raster – a hand-crafted fusion of NLCD land-cover and OSM layers (PROC-STACK)
• Splits: Train = Pittsburgh; OOD validation/test = Austin & Durham. Auxiliary layers raise OOD overall accuracy by ~4 pp without extra fine-tuning.

---

BigEarthNet v2.0 – Multi-Label Land-Cover Classification

• Optical input: 10-band Sentinel-2 tile pairs; ≈ 550 k patch/label pairs over 19 classes.
• Auxiliary layer:
  • SatCLIP location embedding (256-D), one per image center, injected as an extra ViT token (TOKEN-FUSE).
• Splits: Grid-based; val/test tiles lie outside the training footprint (spatial OOD by design). SatCLIP token lifts macro-F1 by ~3 pp across all subset sizes.

---

USAVars – Tree-Cover Regression

• Optical input: NAIP RGB-NIR images (1 km² tiles); ≈ 100 k samples with tree-cover % labels.
• Auxiliary layers:
  • Extended OSM raster stack (roads, buildings, land-use, biome classes, …)
  • Continental Europe Digital Elevation Model (DEM) resampled to 10 m GSD
• Notes: Stacking the OSM raster boosts R² by 0.16 in the low-data regime (< 250 images); DEM is provided raw for flexibility.

Usage Instructions

• Download the .h5.gz files in data/<source dataset name>. Our source datasets include SustainBench, USAVars, and BigEarthNet2.0
• You may use pigz (https://linux.die.net/man/1/pigz) to decompress the archive. This is especially recommended for USAVars' train-split, which is 117 GB when uncompressed. This can be done with pigz -d <.h5.gz>
• Datasets with auxiliary geographic inputs can be read with H5PY.

---