isp-uv-es/worldstrat-taco

WorldStrat

This is a repackaging, not a new dataset. It is WorldStrat by EPFL ECEO (Cornebise, Orsolic, Kalaitzis), Airbus Defence and Space (SPOT 6/7), ESA Copernicus Sentinel-2, converted to TACO. Pixel values and labels are kept as released except where the description below says otherwise. All credit belongs to the original authors: if you use it, please cite them and follow their licence.

original dataset · paper · licence: CC-BY-NC-4.0, CC-BY-4.0

Repackaged into TACO by the Image and Signal Processing Group (ISP), Universitat de València, within the ELLIOT project.

Citation

Please cite the original work:

@inproceedings{cornebise2022open,
  title     = {Open High-Resolution Satellite Imagery: The WorldStrat Dataset -- With Application to Super-Resolution},
  author    = {Cornebise, Julien and Or{\v{s}}oli{\'c}, Ivan and Kalaitzis, Freddie},
  booktitle = {Advances in Neural Information Processing Systems 35 (NeurIPS 2022) Datasets and Benchmarks Track},
  year      = {2022}
}

About the data

3890 globally stratified 2.5 km AOIs, each pairing a SPOT 6/7 high-resolution acquisition (1.5 m panchromatic and a 6 m four-band cube) with a Sentinel-2 L2A series of 16 revisits over the same ground. Every revisit carries its acquisition date, cloud cover, cloud mask and signed day offset from the HR date. The target is an image rather than a label.

3,890 samples · splits: test 389 · train 3,109 · validation 392 · tasks: super-resolution

Packaged as TACO v3.

Full description

Sampling. Stratified over IPCC land cover and settlement density, over-sampling ASMSpotter, Amnesty, Landcover, UNHCR sites, so the corpus is globally representative by construction rather than by availability.

Georeferencing. The release ships an identity transform tagged EPSG:4326; the transform is rebuilt from the published bounds, giving 1.50 / 6.00 / ~10 m per pixel. A stray non-AOI directory in hr_dataset is also dropped.

Products. The pan-sharpened and PNG products are dropped as derivations of what is kept. Sentinel-2 is physical float reflectance in 0..1 rather than scaled uint16.

Splits. The official stratified split is preserved: 3109 train, 392 val, 389 test. 38 AOIs on disk are excluded because that split does not cover them.

Licence. The SPOT 6/7 high-resolution imagery is CC-BY-NC-4.0: Zenodo 6810792 declares licence id other-nc and states the Airbus imagery is 'distributed, with authorization from Airbus, under Creative Commons Attribution-NonCommercial 4.0 International'. Only the labels and the Sentinel-2 half are CC-BY-4.0, so the whole container is non-commercial.

Getting started

git clone --recursive https://github.com/OscarPellicer/taco
pip install -e "taco/python[ml]"     # builds the reader: C++23, CMake, Ninja, pkg-config, libcurl >= 7.83, OpenSSL >= 3

Read it straight from the Hub:

from huggingface_hub import snapshot_download
from taco.ml import Dataset, plot_sample

path = snapshot_download("isp-uv-es/worldstrat-taco", repo_type="dataset")
ds = Dataset(path)                  # the repository is the container: parts + .tacocat
plot_sample(ds[0])

or from a local copy:

ds = Dataset("worldstrat")
sample = ds[0]                      # {slot name: SlotValue}, arrays decoded
sample["lr"].array.shape

Metadata without decoding anything:

import taco
taco.read("worldstrat/.tacocat")              # one Arrow table, levels joined

Samples

What a sample contains

role	slot	holds	modality	detail
input	lr	raster_series	multispectral	12 band(s), unit 1, physical
input	lr_cloud	mask_set	label_raster	2 classes
target	hr_pan	raster	optical	1 band(s), digital_number
target	hr_ms	raster	optical	4 band(s), digital_number

Licence

CC-BY-NC-4.0, CC-BY-4.0

Terms of use:

• Non-commercial use only (CC-BY-NC-4.0). This applies to the whole dataset, including its Sentinel-2 part.

Required credits:

• WorldStrat: Cornebise, Orsolic and Kalaitzis
• SPOT 6/7 imagery © Airbus DS
• Contains modified Copernicus Sentinel data

Providers: EPFL ECEO (Cornebise, Orsolic, Kalaitzis), Airbus Defence and Space (SPOT 6/7), ESA Copernicus Sentinel-2

Acknowledgements

TACO was designed by César Aybar and is specified at https://asterisk.coop/taco/spec/.

Built by Oscar Pellicer within the ELLIOT project at the Image and Signal Processing Group (ISP), Universitat de València.