UT Austin Geoscience Hackathon: `AlphaEarthHack` team
VectorScope: Similarity Search with Google’s Satellite Embedding V1 empowers users to map surface types by drawing sample points or loading point assets. The app computes cosine-similarity heatmaps of 64-dimensional embeddings for any year, offers threshold tuning and heatmap toggling among other functionalities
Binary classification experiments to interpret Google AlphaEarth Foundation embeddings across ESA WorldCover land cover classes. Part of the study "What on Earth is AlphaEarth?" — 130,000+ experiments using Random Forest, XGBoost, LightGBM and progressive ablation across 64 embedding dimensions.
Embedding-based flood detection using AlphaEarth satellite embeddings, GeoTIFF flood datasets, and machine learning for scalable geospatial flood classification.
A unified Python interface to Earth embeddings: image encoders, location encoders, and precomputed geospatial products.
Desert-vegetation boundary detection by projecting self-supervised satellite embeddings onto a vegetation-aligned axis. IGARSS 2026.
An end-to-end GeoAI platform for agricultural monitoring using satellite foundation model embeddings, AlphaEarth, Galileo representations, NDVI-based vegetation analysis, and machine learning based crop classification.
Near-term wildfire risk forecasting platform for California using H3 geospatial indexing, NASA FIRMS, AlphaEarth embeddings, weather features, baseline ML models, and a map-ready dashboard.
Unsupervised agricultural anomaly mapping using AlphaEarth satellite embeddings, K-Means baseline clustering, and CHIRPS rainfall validation.
AI-powered geospatial similarity search using Google DeepMind AlphaEarth Foundation and Google Earth Engine to identify potential iron ore mining locations from satellite imagery.
SRS submission package for frozen AlphaEarth embeddings in bounded LULC change screening and allocation diagnostics
Yearly AlphaEarth embeddings for the 671 CAMELS-US basins (2017-2025), as a Python package + Zenodo dataset.
Browser RGB viewer for AlphaEarth's 64-D satellite embeddings — no login, self-hosted (gray-code band scrub + TiTiler)
Ask the planet a question. Natural-language vector search over satellite imagery embeddings: type 'solar farms in deserts' and get matching locations on a map. Built on AlphaEarth embeddings and EmbeddingGemma 2.
Species distribution modeling at up to 10m resolution on Google's AlphaEarth satellite embeddings. Data extraction, model training and mapping all run server-side on Google Earth Engine.
中国 2017–2024 年 10 m 土地覆盖生产（AlphaEarth Foundations 年度嵌入 + GEE 服务端随机森林，24 类）｜Annual 10 m land cover for China (2017–2024), AEF embeddings + GEE Random Forest
Real-data GeoAI investigation workbench: AlphaEarth landscape change, Sentinel-2 evidence, similarity search, spatial evaluation, and GIS exports.
To associate your repository with the alphaearth topic, visit your repo's landing page and select "manage topics."