AlphaEarth Foundations

A "virtual satellite" to map Earth in unprecedented detail

91 followers

A "virtual satellite" to map Earth in unprecedented detail

91 followers

New AI model integrates petabytes of Earth observation data that stitches together data from actual satellite images, radar, climate simulations, and more to map Earth’s land and coastal waters.

Free

Launch tags:Space•Climate Tech•Data Science

Launch Team

Webflow | AI site builder — Start fast. Build right.

Start fast. Build right.

Promoted

Raycast

Hunter

📌

This is a super cool new model from Google DeepMind!

We’re introducing a new way to analyze the planet. Google’s Satellite Embedding dataset uses the power of AI to pack a year’s worth of multi-source satellite data into every single 10-meter pixel, enabling faster and more powerful geospatial analysis. Welcome to the future of deep learning in Earth Engine.

The model enables a kind of "virtual satellite" that can be used to see what's going on all over the planet — in 10m increments.

This first-of-its-kind dataset was generated using AlphaEarth Foundations, Google DeepMind’s new geospatial AI model that assimilates observations across diverse sources of geospatial information, including optical and thermal imagery from Sentinel-2 and Landsat satellites, radar data that can see through clouds, 3D measurements of surface properties, global elevation models, climate information, gravity fields, and descriptive text. Unlike traditional deep learning models that require users to fine-tune weights and run their own inference on clusters of high-end computers, AlphaEarth Foundations was designed to produce information-rich, 64-dimensional geospatial “embeddings” that are suitable for use with Earth Engine’s built-in machine learning classifiers and other pixel-based analysis.

Report

7mo ago

Haimeta

Wow, pulling together all those different data sources into one crazy-detailed Earth map is genius, tbh. This is gonna help so much with climate stuff, realy impressive work!

Report

7mo ago

Unveil Engine

Read the article, Similarity Search looks looks so cool.

Report

7mo ago

How wonderful. I wonder if I could revisit the hometown of my childhood, to see what it has become？The very thought is intriguing.

Report

7mo ago

AltPage.ai

No way, this is wild—real-time, detailed mapping from so many data sources? As someone who geeks out over coastal changes, I’m genuinely impressed. Can it track seasonal shoreline shifts too?

Report

7mo ago

GPT-4o

This “virtual satellite” idea is wild—being able to map the Earth in that much detail is a total gamechanger tbh. DeepMind team is on another level, fr.

Report

7mo ago

This is a super cool new model from Google DeepMind!

We’re introducing a new way to analyze the planet. Google’s Satellite Embedding dataset uses the power of AI to pack a year’s worth of multi-source satellite data into every single 10-meter pixel, enabling faster and more powerful geospatial analysis. Welcome to the future of deep learning in Earth Engine.

The model enables a kind of "virtual satellite" that can be used to see what's going on all over the planet — in 10m increments.

This first-of-its-kind dataset was generated using AlphaEarth Foundations, Google DeepMind’s new geospatial AI model that assimilates observations across diverse sources of geospatial information, including optical and thermal imagery from Sentinel-2 and Landsat satellites, radar data that can see through clouds, 3D measurements of surface properties, global elevation models, climate information, gravity fields, and descriptive text. Unlike traditional deep learning models that require users to fine-tune weights and run their own inference on clusters of high-end computers, AlphaEarth Foundations was designed to produce information-rich, 64-dimensional geospatial “embeddings” that are suitable for use with Earth Engine’s built-in machine learning classifiers and other pixel-based analysis.