Upload GeoFlood-275 dataset

.gitattributes

@@ -58,3 +58,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+metadata/sample_index_train(1).csv filter=lfs diff=lfs merge=lfs -text
+metadata/sample_index_train.csv filter=lfs diff=lfs merge=lfs -text

CITATION(1).bib

@@ -0,0 +1,6 @@
+@misc{li2026geoflood275,
+  title = {Toward Rapid Flood Mapping Anywhere via Terrain- and Land-Cover-Conditioned Optical-SAR Fusion},
+  author = {Li, Jiepan and Huang, He and Li, Wenke and Li, Linxin and Xie, Anqi and Ye, Ruoru and Hu, Lei and Hu, Ting and He, Wei and Zhang, Liangpei},
+  year = {2026},
+  note = {Manuscript under revision for Remote Sensing of Environment}
+}

LICENSE(1)

@@ -0,0 +1,5 @@
+GeoFlood-275 is released under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
+
+License summary and legal code: https://creativecommons.org/licenses/by/4.0/
+
+Third-party source products used to construct the benchmark retain their original terms, notices, and attribution requirements. See NOTICE.md for source-data attribution.

NOTICE(1).md

@@ -0,0 +1,16 @@
+# Notice
+
+GeoFlood-275 is a derived benchmark constructed from public geospatial products and event metadata. The benchmark packaging is released under CC BY 4.0. Third-party source data retain their original terms, notices, and attribution requirements.
+
+Source products used in GeoFlood-275 include:
+
+- Copernicus Emergency Management Service Rapid Mapping EMSR delineation products
+- Copernicus Sentinel-1 GRD imagery
+- Copernicus Sentinel-2 Level-1C imagery
+- Copernicus DEM GLO-30
+- ESA WorldCover v200 2021
+- Dynamic World V1/NRT land-cover labels
+
+GeoFlood-275 reference labels are derived from EMSR delineation vectors and are intended for research benchmarking of event-induced flood inundation mapping. They should not be treated as authoritative emergency-response products or absolute hydrodynamic ground truth.
+
+When using this dataset, cite the GeoFlood-275 paper and acknowledge the source data providers.

README(1).md

@@ -0,0 +1,184 @@
+---
+pretty_name: GeoFlood-275
+license: cc-by-4.0
+task_categories:
+- image-segmentation
+language:
+- en
+tags:
+- remote-sensing
+- flood-mapping
+- flood-inundation
+- sentinel-1
+- sentinel-2
+- sar
+- optical-sar-fusion
+- terrain
+- land-cover
+- geotiff
+- webdataset
+size_categories:
+- 100K<n<1M
+---
+
+# GeoFlood-275
+
+GeoFlood-275 is a global event-level benchmark for rapid event-induced flood inundation mapping from heterogeneous optical-SAR observations with terrain and land-cover information. It contains 275 flood event-AOI pairs worldwide and 133,555 512 x 512 GeoTIFF patch samples.
+
+The benchmark follows an operational setting: pre-event Sentinel-2 optical imagery provides antecedent land-surface context, post-event Sentinel-1 SAR imagery provides all-weather crisis observations, and slope plus land-cover layers provide environmental context. Reference inundation maps are derived from Copernicus Emergency Management Service Rapid Mapping (EMSR) delineation products and represent event-induced inundation within the official EMSR AOI. Permanent or pre-existing surface-water bodies are not treated as target flood pixels under the adopted EMSR delineation definition.
+
+This repository packages the benchmark as WebDataset TAR shards to keep the public Hugging Face dataset efficient to browse, download, and stream.
+
+## Dataset Paper
+
+**Toward Rapid Flood Mapping Anywhere via Terrain- and Land-Cover-Conditioned Optical-SAR Fusion**
+
+Jiepan Li, He Huang, Wenke Li, Linxin Li, Anqi Xie, Ruoru Ye, Lei Hu, Ting Hu, Wei He, and Liangpei Zhang.
+
+Manuscript under revision for *Remote Sensing of Environment*.
+
+## Splits
+
+GeoFlood-275 uses event-level temporal separation. Events from 2015-2025 are used for training/model development; events from 2026 onward are reserved for temporally separated validation and testing.
+
+| split | event-AOI pairs | patch generation | samples | shards | payload |
+| --- | ---: | --- | ---: | ---: | ---: |
+| train | 234 | 512 x 512 patches, stride 256 | 125,552 | 86 | 342.57 GiB |
+| validation | 20 | 512 x 512 patches, stride 512 | 4,476 | 3 | 11.92 GiB |
+| test | 21 | 512 x 512 patches, stride 512 | 3,527 | 3 | 9.46 GiB |
+| total | 275 | - | 133,555 | 92 | 363.96 GiB |
+
+Patches with more than 70% invalid pixels were discarded during construction. Boundary patches are padded when necessary.
+
+## Repository Layout
+
+```text
+README.md
+LICENSE
+NOTICE.md
+CITATION.bib
+data/
+  train/geoflood-train-000000.tar
+  validation/geoflood-validation-000000.tar
+  test/geoflood-test-000000.tar
+metadata/
+  shard_manifest.jsonl
+  sample_index_train.csv
+  sample_index_validation.csv
+  sample_index_test.csv
+```
+
+Each WebDataset sample contains files sharing the same sample key:
+
+```text
+<sample_key>.s2.tif
+<sample_key>.s1.tif
+<sample_key>.s1pre.tif
+<sample_key>.flood.tif
+<sample_key>.dem.tif
+<sample_key>.dynamic_landcover.tif
+<sample_key>.esav200.tif
+<sample_key>.slope_norm.tif
+<sample_key>.json
+```
+
+Example sample key:
+
+```text
+EMSR251_02ARENDAL_DEL_MONIT01_v2_0_0
+```
+
+## Modalities
+
+| suffix | source layer | bands | dtype | no-data | notes |
+| --- | --- | ---: | --- | --- | --- |
+| `s2.tif` | pre-event Sentinel-2 Level-1C | 4 | Float32 | -100 | B4, B3, B2, B8; stored in clipped reflectance scale, typically 0-6000 |
+| `s1.tif` | post-event Sentinel-1 GRD | 2 | Float32 | -50 | VV and VH backscatter in dB |
+| `s1pre.tif` | auxiliary pre-event Sentinel-1 GRD | 2 | Float32 | -50 | optional layer for SAR-SAR or optical+SAR ablation experiments |
+| `flood.tif` | EMSR-derived inundation reference | 1 | Byte | none | binary event-induced flood mask, 1 = flooded, 0 = non-flood |
+| `dem.tif` | Copernicus DEM GLO-30 | 1 | Float32 | -9999 | elevation layer aligned to Sentinel-1 reference grid |
+| `dynamic_landcover.tif` | Dynamic World NRT land-cover composite | 1 | Int16 | 0 | event-adjacent 14-day pre-event majority-vote land-cover labels |
+| `esav200.tif` | ESA WorldCover v200 2021 | 1 | Int16 | 0 | static 10 m land-cover reference |
+| `slope_norm.tif` | normalized slope derived from Copernicus DEM | 1 | Float32 | 0 | slope clipped to 0-45 degrees and normalized to 0-1 |
+| `json` | sample metadata | - | JSON | - | split, source event id, tile coordinates, and original relative paths |
+
+The released GeoTIFF files are not pre-normalized training tensors. They preserve the benchmark raster values; normalization is applied in the training dataloader. The reference implementation is `utils/dataloader.py` in the GeoFloodNet/RapidFloodMapping codebase: Sentinel-1 is clipped to [-45, 25] dB and mapped to [-1, 1], Sentinel-2 is clipped to [0, 6000] and mapped to [-1, 1], DEM is scaled by `/5000` then standardized with mean 0.5 and std 0.5, slope is clipped to [0, 1], ESA WorldCover raw codes are remapped to contiguous IDs, Dynamic World labels are kept as integer labels, and flood masks are mapped to {0, 1} when needed.
+
+## Loading
+
+Install the common tooling:
+
+```bash
+python -m pip install -U datasets huggingface_hub rasterio webdataset
+```
+
+Stream with Hugging Face Datasets:
+
+```python
+from datasets import load_dataset
+
+repo_id = "jiepanli/GeoFlood-275"
+data_files = {
+    "train": f"hf://datasets/{repo_id}/data/train/*.tar",
+    "validation": f"hf://datasets/{repo_id}/data/validation/*.tar",
+    "test": f"hf://datasets/{repo_id}/data/test/*.tar",
+}
+
+ds = load_dataset("webdataset", data_files=data_files, streaming=True)
+sample = next(iter(ds["train"]))
+print(sample.keys())
+```
+
+Download selected shards and read GeoTIFF bytes with Rasterio:
+
+```python
+from glob import glob
+from huggingface_hub import snapshot_download
+from rasterio.io import MemoryFile
+import webdataset as wds
+
+repo_id = "jiepanli/GeoFlood-275"
+root = snapshot_download(
+    repo_id=repo_id,
+    repo_type="dataset",
+    allow_patterns=["data/test/*.tar", "metadata/*", "README.md", "NOTICE.md", "CITATION.bib"],
+)
+
+urls = sorted(glob(f"{root}/data/test/*.tar"))
+dataset = wds.WebDataset(urls)
+
+sample = next(iter(dataset))
+with MemoryFile(sample["s2.tif"]) as memfile:
+    with memfile.open() as src:
+        s2 = src.read()  # shape: bands x height x width
+
+print(sample["__key__"], s2.shape)
+```
+
+## Source Data And Attribution
+
+GeoFlood-275 is built from independently retrieved and processed public geospatial products anchored by EMSR event metadata:
+
+- Copernicus Emergency Management Service Rapid Mapping EMSR delineation products: https://emergency.copernicus.eu/mapping/ems/rapid-mapping
+- Sentinel-2 Level-1C imagery from `COPERNICUS/S2`: https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S2
+- Sentinel-1 GRD imagery from `COPERNICUS/S1_GRD`: https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_S1_GRD
+- Copernicus DEM GLO-30: https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_DEM_GLO30
+- ESA WorldCover v200 2021: https://developers.google.com/earth-engine/datasets/catalog/ESA_WorldCover_v200
+- Dynamic World V1/NRT land-cover labels: https://developers.google.com/earth-engine/datasets/catalog/GOOGLE_DYNAMICWORLD_V1
+
+Please cite the GeoFlood-275 paper and acknowledge the above source products when using this dataset. Users are responsible for complying with the original source-data terms in addition to the GeoFlood-275 license.
+
+## Limitations
+
+The EMSR-derived references are expert-interpreted operational products, not absolute hydrodynamic ground truth. Residual clouds and cloud shadows may remain in some pre-event Sentinel-2 observations because the EMSR-designated pre-event acquisition is retained rather than replaced by an independently selected cloud-free image. ESA WorldCover and Dynamic World are useful but imperfect land-cover references and should not be interpreted as deterministic flood/non-flood constraints.
+
+## Citation
+
+```bibtex
+@misc{li2026geoflood275,
+  title = {Toward Rapid Flood Mapping Anywhere via Terrain- and Land-Cover-Conditioned Optical-SAR Fusion},
+  author = {Li, Jiepan and Huang, He and Li, Wenke and Li, Linxin and Xie, Anqi and Ye, Ruoru and Hu, Lei and Hu, Ting and He, Wei and Zhang, Liangpei},
+  year = {2026},
+  note = {Manuscript under revision for Remote Sensing of Environment}
+}
+```

data/test/geoflood-test-000000.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ee2ca5bff5e1d5399caa5918dccd6197d17c29f6bacf0bffca7531c87b663a8
+size 4306350080

data/test/geoflood-test-000000.tar.done(1).json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e62bfd6d90599ea130c615cfd09b6c1767b05c52cb15a69ce8f3a69328896e4
+size 269

data/test/geoflood-test-000000.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e62bfd6d90599ea130c615cfd09b6c1767b05c52cb15a69ce8f3a69328896e4
+size 269

data/test/geoflood-test-000001.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e5103db8953de7204c56ea2ba9975b02d350da7f9ebf764f8a59efc293baf20
+size 4298946560

data/test/geoflood-test-000001.tar.done(1).json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb1e2f125c3d58084ca95fcae12eaac2aaf97559f3010703698f8717ca45c97d
+size 272

data/test/geoflood-test-000001.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb1e2f125c3d58084ca95fcae12eaac2aaf97559f3010703698f8717ca45c97d
+size 272

data/test/geoflood-test-000002.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a6969a661e50c715017aab44b6f395e5d4d119066c1b3ea3f627beef97e267a
+size 1583749120

data/test/geoflood-test-000002.tar.done(1).json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11bb8e26ca4df727ed07bbb4f347f4314d860ddd15dfbd372a035373d4d491cb
+size 273

data/test/geoflood-test-000002.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11bb8e26ca4df727ed07bbb4f347f4314d860ddd15dfbd372a035373d4d491cb
+size 273

data/train/geoflood-train-000000.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4442ce69cbf4f1613b43abcc73d8ddcef8ff131c02d67ab27da7d4335997d0d0
+size 4300687360

data/train/geoflood-train-000000.tar.done(1).json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a4014bf132a9dfcc28097341b99e5a009e344744e6b46ed304ff5b72b36aa8
+size 283

data/train/geoflood-train-000000.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a4014bf132a9dfcc28097341b99e5a009e344744e6b46ed304ff5b72b36aa8
+size 283

data/train/geoflood-train-000001.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b86f16908050e26bffc48a26a8ef0817935fa164a832ff73689f926847470553
+size 4302592000

data/train/geoflood-train-000001.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cbcc05334ecdf6576f788d53c9aff50b096fb68cd5bc61e3d4a3f3c8b781faa
+size 278

data/train/geoflood-train-000002.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eca7e045263c99021ac4052e64d80377e19e1b09c592f32629f74139d0546aef
+size 4299130880

data/train/geoflood-train-000002.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:597aefc66ec97cbb105602d3838cbbb9ea61a5028198850c68b0fc2b15ff4920
+size 283

data/train/geoflood-train-000003.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:369a8685593a808c457f5806f3ad911bd30d923dd1f589394c83f2bacf12eed4
+size 4301834240

data/train/geoflood-train-000003.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c652ab7e63d8e9d3f2a8c010b833e8c45daabe9fe6715062dee03a5f8c89632
+size 290

data/train/geoflood-train-000004.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cef7c1d6820971a33a1858d4326e60f7c9e28a499d0b9e8e4007baab0420319f
+size 4300574720

data/train/geoflood-train-000004.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3496eb60576b946d4e870242788dd305a34439227c2735d1db88e4043c085915
+size 286

data/train/geoflood-train-000005.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:452da8d6ac33dbf98848c854e3d5c0d34ad4f889ee5a487b2f3e7d651d20d68c
+size 4297840640

data/train/geoflood-train-000005.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2d5b631a4bd81df5eb24d1ed2682fed4438182241c45414ddda370206b86c99
+size 283

data/train/geoflood-train-000006.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ac894611e83063a8f95f0db8a3e7ef377bf45b327060cb2bbb78c5b065864b2
+size 4298393600

data/train/geoflood-train-000006.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d06cc8ddb92e50b8b802ad8fa6c826f9672a5640a76586e9cc78b1a28de928e3
+size 278

data/train/geoflood-train-000007.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83092e2968fb954aa72f56cf35ba122eab3930a6e06b53f9730313ff76135633
+size 4297891840

data/train/geoflood-train-000007.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f52c9a7bce446a5dd2af2405200f4322d125cf591a3b4e3dbcc535de3fb6787d
+size 279

data/train/geoflood-train-000008.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2031a08f96bedbd6ebbe4dc98bfdd60473070605cf4979031bb2653439eb3ea5
+size 4301312000

data/train/geoflood-train-000008.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d80761fe711060f1788e85e45866fed94b4b3d48272618de281a4b2e7331520a
+size 281

data/train/geoflood-train-000009.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8729e2108d573a067664d1e18d11744e119ba68e4b054acc3d9edf8dda82a940
+size 4298076160

data/train/geoflood-train-000009.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c60867f05fcef0af374a814b1aed6756fbe5bec16e62aa8159b12cc56cbbbc09
+size 273

data/train/geoflood-train-000010.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d9e9c2737b3174af89c71f4f66e14509eb0575a3df0c6660a695a2d6c0cc087
+size 4300318720

data/train/geoflood-train-000010.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:543df07ce7b4d27d45865e6097c12d35d41f7dac7502bdea1c6a27c96fd89d50
+size 269

data/train/geoflood-train-000011.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e501bd076b0db1b8fb0a0d6807b27d5534a5f564d1de5c5ca00fab4ad2c425e
+size 4297543680

data/train/geoflood-train-000011.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:552113b70d78d5eca0ebb457bbf4b9cd9cb0b80eddf1c93bbbbc7cf8542a3b09
+size 274

data/train/geoflood-train-000012.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e42779dc67e6568dc8805a9165e938789076ffcaecff276e0af824dd1c3b5edc
+size 4298024960

data/train/geoflood-train-000012.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de725aee6460affe01362e46254f909dc5d427e5860e417d755a289b9e87ed92
+size 279

data/train/geoflood-train-000013.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbaf3eb266a2d578a2b9854efce25b8b4d326ef84f8cbd4b9f7d51ea5906a919
+size 4298373120

data/train/geoflood-train-000013.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9dbb41ceeebafdbf3acd44d39d7f76bbee2d110bfaf0544564882a2b07f17cf
+size 272

data/train/geoflood-train-000014.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:276f7aeada614738d11a0ee8757be0a83c1d435dd74a0b6e657d758093dfc957
+size 4301199360

data/train/geoflood-train-000014.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76ad928290f37039782838d76ffe1c7fc25bd0f91222f716d87ae75e764f97c3
+size 270

data/train/geoflood-train-000015.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0c8a18a0570545564be50f3d26eb5a53c89770b64465e2a52409df684622235
+size 4302520320

data/train/geoflood-train-000015.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed92dbaf27719989e0a294bf8e33f827b71e6949f05c77f2785909da7c0d3c50
+size 270

data/train/geoflood-train-000016.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a3f4e83c193a228659b2dde2714ab3b2c7b17af10de8eb4201a26a560067407
+size 4297205760

data/train/geoflood-train-000016.tar.done.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19141ae3f1d9054e22f82b048b37e6ee3b9d60121bc8c6d77a930187dc9f8001
+size 271

data/train/geoflood-train-000017.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c91f0645ecc132bbd3475676f955e09d82639b2a5b9c102a761cc95ef54607d
+size 4295843840