A neutral facial expression of a wearer of a head-mountable display (HMD) is detected based on signals received from one or multiple facial electromyographic (fEMG) sensors of the HMD. A baseline neutral facial expression image of the wearer is responsively automatedly captured using one or multiple cameras of the HMD. HMD camera-based detection of facial expressions of the wearer is calibrated based on the automatedly captured baseline neutral facial expression image.