TerraMesh

TerraMesh-Masks TerraMesh-Masks is a dataset for open-vocabulary segmentation of satellite imagery. This dataset provides binary segmentation masks with captions that extend the samples from TerraMesh. We also provide an human-verfied evaluation benchmark, called TerraMesh-Masks-Eval. Examples from the training subset: Usage Download the data loading code from GitHub and install requirements with pip install -r requirements.txt. For development, you can… See the full description on the dataset page: https://huggingface.co/datasets/ibm-esa-geospatial/TerraMesh-Masks.

TerraMesh-Masks-Eval TerraMesh-Masks-Eval is a human-verified benchmark dataset to evaluate open-vocabulary segmentation models on satellite imagery. This dataset provides binary segmentation masks with captions togehther with input samples from TerraMesh. We also provide a training dataset, called TerraMesh-Masks. Examples from the evaluation subset: Usage Download the data loading code from GitHub and install requirements with pip install -r… See the full description on the dataset page: https://huggingface.co/datasets/ibm-esa-geospatial/TerraMesh-Masks-Eval.

Dataset Card for Multi-Temporal Crop Classification Dataset Summary This dataset contains temporal Harmonized Landsat-Sentinel imagery of diverse land cover and crop type classes across the Contiguous United States for the year 2022. The target labels are derived from USDA's Crop Data Layer (CDL). It's primary purpose is for training segmentation geospatial machine learning models. Dataset Structure TIFF Files Each tiff file covers a… See the full description on the dataset page: https://huggingface.co/datasets/ibm-nasa-geospatial/multi-temporal-crop-classification.

Time series of annual global maps of land use and land cover (LULC). It currently has data from 2017-2023. The maps are derived from ESA Sentinel-2 imagery at 10m resolution. Each map is a composite of LULC predictions for 9 classes throughout the year in order to generate a representative snapshot of each year. This dataset, produced by Impact Observatory, Microsoft, and Esri, displays a global map of land use and land cover (LULC) derived from ESA Sentinel-2 imagery at 10 meter resolution for the years 2017 - 2023. Each map is a composite of LULC predictions for 9 classes throughout the year in order to generate a representative snapshot of each year. This dataset was generated by Impact Observatory, which used billions of human-labeled pixels (curated by the National Geographic Society) to train a deep learning model for land classification. Each global map was produced by applying this model to the Sentinel-2 annual scene collections from the Mircosoft Planetary Computer. Each of the maps has an assessed average accuracy of over 75%. These maps have been improved from Impact Observatory’s previous release and provide a relative reduction in the amount of anomalous change between classes, particularly between “Bare” and any of the vegetative classes “Trees,” “Crops,” “Flooded Vegetation,” and “Rangeland”. This updated time series of annual global maps is also re-aligned to match the ESA UTM tiling grid for Sentinel-2 imagery. All years are available under a Creative Commons BY-4.0.

Global, cloud-free, and reconstructed historical Landsat spectral bands are provided at a spatial resolution of 30 meters and temporal resolution of bimonthly intervals from 1997 onwards. The dataset was generated using the GLAD Landsat ARD version 2 as the primary input for temporal aggregation and the imputation of missing values.

Annual Earth-embedding layers from Google DeepMind's AlphaEarth Foundations model: for every 10 m square of land, 64 numbers that summarise what the satellite record (optical, radar, elevation, climate and more) shows there over the year. Places that look and behave alike have similar embeddings, so they power similarity search (“find places like this”), land-cover mapping and change detection without hand-labelled training data. Coverage: global land and coastal waters, one layer per year from 2017 to 2025, in UTM tiles of 8192 × 8192 pixels.