Measurements of the star formation activity on cloud scales are fundamental to uncovering the physics of the molecular cloud, star formation, and stellar feedback cycle in galaxies. Infrared (IR) emission from small dust grains and polycyclic aromatic hydrocarbons (PAHs) is widely used to trace the obscured component of star formation. However, the relation between these emission features and dust attenuation is complicated by the combined effects of dust heating from old stellar populations and an uncertain dust geometry with respect to heating sources. We used images obtained with NIRCam and MIRI as part of the PHANGS–JWST survey to calibrate the IR emission at 21 μm, and the emission in the PAH-tracing bands at 3.3, 7.7, 10, and 11.3 μm as tracers of obscured star formation. We analysed ∼20 000 optically selected H II regions across 19 nearby star-forming galaxies, and benchmarked their IR emission against dust attenuation measured from the Balmer decrement. We modelled the extinction-corrected Hα flux as the sum of the observed Hα emission and a term proportional to the IR emission, with a_IR as the proportionality coefficient. A constant a_IR leads to an extinction-corrected Hα estimate that agrees with those obtained with the Balmer decrement with a scatter of ∼0.1 dex for all bands considered. Among these bands, 21 μm emission is demonstrated to be the best tracer of dust attenuation. The PAH-tracing bands underestimate the correction for bright H II regions, since in these environments the ratio of PAH-tracing bands to 21 μm decreases, signalling destruction of the PAH molecules. For fainter H II regions, all bands suffer from an increasing contamination from the diffuse IR background. We present calibrations that take this effect into account by adding an explicit dependence on 2 μm emission or stellar mass surface density.