This is not my understanding, I have read at least 7 or 8 papers that seem to have for various models and techniques reduced the delta between masked and unmasked recognition to be very similar on false positive and positive rates.

https://www.mdpi.com/2076-3417/11/16/7310 https://www.sciencedirect.com/science/article/pii/S240589632... ...

That said I have no insight as to how many of these techniques have been found to scale well or have started to make it into product. It has been publicly reported that NEC’s NeoFace (a system that many police and govt use) newer versions does indeed have occlusion (mask) recognition operating at very high levels.

anyways thats just my understanding as an interested bystander -- not in the field.