synapse-sr 0.3.0

Sentinel-2 at 2 m, with every pixel accounted for

synapse-sr turns a Sentinel-2 L2A scene into a 2.0 m red / green / blue / near-infrared GeoTIFF. A physical model of the instrument pins everything the satellite measured. The network only adds what 10 m pixels cannot show, and every output says which pixels came from the measurement and which came from the learned prior.

pip install synapse-sr

import synapse_sr

r = synapse_sr.super_resolve("sentinel2_l2a.tif")   # progress bar, then a 5x larger result
r.save("sentinel2_2m.tif")                           # georeferenced, same CRS and bounds
r.summary()                                          # size, consistency, support, time

synapse-sr sentinel2_l2a.tif sentinel2_2m.tif        # the same from the shell

What it is for

Worked examples for each: Applications.

Pro or Flash

Both share the same observation-consistent pipeline, so the same guarantees hold for either one. Flash weights are not released yet. Until they are, use Pro, which also runs on CPU (more slowly). synapse-sr --models shows what is published.

Why trust the output

r.x_base        # what the 10 m observation determines
r.prior         # what the network added (x_base + prior == image), invisible to the sensor
r.support       # per pixel: 2 observation-determined, 1 medium, 0 prior-dominated or invalid
r.uncertainty() # calibrated expected error per pixel and band
r.consistency   # re-observing the output reproduces the input, in sensor-noise units

x_hat = x_base + P_N(delta): P_N removes every component the sensor could have seen from the network's output, so the network cannot contradict the measurement. See How it works.

Runs everywhere

synapse-sr --env shows what is active on your machine.

Documentation

Quick start · Colab and Kaggle · Applications · Examples · API · Limitations

Limitations, in short. A 2 m grid is not 2 m effective resolution; no effective-resolution figure is claimed for this release. Only B04, B03, B02 and B08 are super-resolved; the 20 m bands are spectral context. The training references are from the United States.

@software{synapse_sr,
  title  = {synapse-sr: observation-consistent super-resolution of Sentinel-2 imagery},
  author = {Naidu, Sharadh},
  year   = {2026},
  url    = {https://github.com/SharadhNaidu/synapse-sr}
}