Prithvi
An open foundation model for reading satellite imagery: floods, fires, crops.
Prithvi is a foundation model trained on Harmonized Landsat and Sentinel-2 (HLS) imagery. The repository shows how to fine-tune it for specific tasks: detecting floods, mapping burn scars, classifying crops across seasons.
And where does AI come in?
A vision transformer trained as a masked autoencoder on HLS data. The encoder weights can be downloaded and the example datasets are published, so that others can build their own Earth observation tools.
The AI label
- Model or technique
- Vision transformer (ViT) trained as a masked autoencoder, with a segmentation head added for each task.
- Model published
- Not filled in
- Where it runs
- Not filled in
- Data
- Harmonized Landsat and Sentinel-2 (HLS) imagery; example datasets on floods, fires and crops.
- Role of people
- Not filled in
- Frugality
- Not filled in
“Not filled in”: MAUB could not check it with the project. Know more? Write to us.
Run this project? Show the “Listed on MAUB” badge
Paste this code on your site: an image and a link to this page. No script, no tracker: MAUB cannot tell who displays it.
<a href="https://maub.subalpine.it/en/projets/prithvi"><img src="https://maub.subalpine.it/statique/img/badge-recense-en.svg" alt="Listed on MAUB" width="180" height="48"></a>Something wrong or out of date? Write to contact@subalpine.it.
The monthly newsletter
Once a month, what moved on MAUB, in your inbox.