We demonstrate real-time magnetometry by detecting single-photon emissions from a nitrogen vacancy center prepared in the setting of coherent population trapping and by estimating magnetic field from the time series of photon counts, which are correlated with the underlying field.  The proof-of-principle experiment uses an external magnetic field that follows an Ornstein-Uhlenbeck (OU) process.  By taking advantage of the statistical properties of the OU process, a Bayesian inference-based estimator can effectively update dynamical information of the field with the detection of just a single photon, enabling sensing of time-varying magnetic fields with single detected photons.