React records tactile images and sensor pose. It has no load cell, so the newtons in it are estimated from the images alone — reconstructed to a depth map, then calibrated on presses of known load.
The only step that needs a choice is image → surface gradient. A lookup table learns it from ~700 sphere presses per sensor; a linear photometric solve needs only the LED geometry. On markerless GelSight Mini presses the calibration-free solve leads by +0.21 ρ; on a marker gel it trails by -0.18, which is what the physics predicts.
Full comparison on the results page · how it works on method.