Between the two of us, we reviewed 22 paper submissions this summer, spread across NeurIPS, WACV, and TerraBytes (a geospatial workshop at ECCV). Fifteen of the 22 (68%) contained entirely fabricated citations,…

Sentinel-2 Paint: Recreate Any Image from Real Satellite Imagery Permalink

Published: July 14, 2026

What colors does the Earth actually come in? We built Sentinel-2 Paint to find out: a script that scours the Sentinel-2 archive for patches of every color the planet shows, and a browser app for playing with the…

Similar but Different: A Benchmark for Measuring Whether Models Actually Use Multispectral Bands Permalink

Published: July 07, 2026

The broad availability of multispectral data is one thing that separates satellite-based machine learning from mainstream computer vision. Sentinel-2 L2A data has 12 surface-reflectance bands, which include visible…

Gaussian Splat-based Satellite Image Super Resolution Permalink

Published: May 12, 2026

Multi-temporal super-resolution of a Sentinel-2 scene. From left: a single S2 natural-color observation (10m), a 0.8m aerial basemap for visual reference, a bicubic 10× upsample of the S2 input, our Gaussian-splat…

ThroughputBench: How fast can a deep learning model map the Earth? Permalink

Published: May 04, 2026

The abundance of open satellite imagery and advances in geospatial ML and remote sensing methods have made it possible to monitor a growing list of variables directly from orbit. Burke et al. (2021, Science), for…

Compressing Earth Embeddings, pt. 3 – DeltaBit Permalink

Published: April 15, 2026

In our TerraBit post last time, we binary-quantized the global Clay v1.5 embeddings down to 128 bytes per patch and served 50M of them from static object storage — implementing planetary scale retrieval entirely in…

Compressing Earth Embeddings Permalink

Published: March 24, 2026

Update (2026-03-26): OlmoEarth-nano results throughout have been recomputed with properly normalized inputs. The initial version we released used unnormalized inputs, which significantly underestimated…

Seeing the Roads Through the Trees: Do Segmentation Models Actually Use Long-Range Context? Permalink

Published: March 17, 2026

How well do segmentation models actually use long-range spatial information to make decisions? No existing benchmark directly measures this, especially in remote sensing where most datasets can be solved with…

Characterizing Census Blocks with Satellite Embedding Statistics Permalink

Published: March 10, 2026

How can you join AEF embeddings to census blocks, and how well do they predict different variables? We wrote a script for doing this! We find, for example, that statistics of AEF embeddings can differentiate between…

Training a Water Segmentation Model with TorchGeo Permalink

Published: March 02, 2026

One notebook, a few hundred lines of Python, and you go from raw Sentinel-2 imagery to a georeferenced water map you can open in QGIS. That’s the premise of the TorchGeo tutorial we put together for the ICLR 2026…

Welcome to GeoSpatial ML Permalink

Published: February 27, 2026

Welcome to GeoSpatial ML — a place to share what we’re exploring, building, and reading at the intersection of geospatial data and machine learning. Many of us already swap papers, datasets, and half-baked…