ABSTRACT
Face recognition is one of the most important research topics for intelligence security system, especially in the COVID-19 era. Medical research has proven that wearing a mask is the most efficient way to avoid the risk of COVID-19. Nevertheless, classic face recognition systems often fail when dealing with masked faces. So it is essential to design a method that is robust to Masked Face Recognition (MFR). In this paper, to relieve the degraded performance of MFR, we propose Mask Aware Network (MAN) including a mask generation module and a loss function searching module. The mask generation module utilizes the face landmarks to obtain more realistic and reliable masked faces for training. The loss function searching module tries to match the most suitable loss for face recognition. On ICCV MFR challenge, our team victor-2021 achieves 5 first places (including 3 champions in standard face recognition and 2 champions in masked face recognition) and 1 third place by 3rd August 2021. These results demonstrate the robustness and generalization of our method in both standard or masked face recognition task.
ABSTRACT
BACKGROUND AND AIM: Coronavirus disease 2019 (COVID-19) has attracted increasing worldwide attention. While diabetes is known to aggravate COVID-19 severity, it is not known whether nondiabetic patients with metabolic dysfunction are also more prone to more severe disease. The association of metabolic associated fatty liver disease (MAFLD) with COVID-19 severity in nondiabetic patients was investigated here. METHODS: The study cohort comprised 65 patients with (i.e. cases) and 65 patients without MAFLD (i.e. controls). Each case was randomly matched with one control by sex (1:1) and age (+/-5 years). The association between the presence of MAFLD (as exposure) and COVID-19 severity (as the outcome) was assessed by binary logistic regression analysis. RESULTS: In nondiabetic patients with COVID-19, the presence of MAFLD was associated with a four-fold increased risk of severe COVID-19;the risk increased with increasing numbers of metabolic risk factors. The association with COVID-19 severity persisted after adjusting for age, sex, and coexisting morbid conditions. CONCLUSION: Health-care professionals caring for nondiabetic patients with COVID-19 should be cognizant of the increased likelihood of severe COVID-19 in patients with MAFLD.