AAN-Face: Attention Augmented Networks for Face Recognition.

Convolutional neural networks are capable of extracting powerful representations for face recognition. However, they tend to suffer from poor generalization due to imbalanced data distributions where a small number of classes are over-represented (e.g. frontal or non-occluded faces) and some of the remaining rarely appear (e.g. profile or heavily occluded faces). This is the reason why the performance is dramatically degraded in minority classes. For example, this issue is serious for recognizing masked faces in the scenario of ongoing pandemic of the COVID-19. In this work, we propose an Attention Augmented Network, called AAN-Face, to handle this issue. First, an attention erasing (AE) scheme is proposed to randomly erase units in attention maps. This well prepares models towards occlusions or pose variations. Second, an attention center loss (ACL) is proposed to learn a center for each attention map, so that the same attention map focuses on the same facial part. Consequently, discriminative facial regions are emphasized, while useless or noisy ones are suppressed. Third, the AE and the ACL are incorporated to form the AAN-Face. Since the discriminative parts are randomly removed by the AE, the ACL is encouraged to learn different attention centers, leading to the localization of diverse and complementary facial parts. Comprehensive experiments on various test datasets, especially on masked faces, demonstrate that our AAN-Face models outperform the state-of-the-art methods, showing the importance and effectiveness.

Predicting Obesity Using Facial Pictures during COVID-19 Pandemic.

BACKGROUND: Sedentary lifestyle and work from home schedules due to the ongoing COVID-19 pandemic in 2020 have caused a significant rise in obesity across adults. With limited visits to the doctors during this period to avoid possible infections, there is currently no way to measure or track obesity. METHODS: We reviewed the literature on relationships between obesity and facial features, in white, black, hispanic-latino, and Korean populations and validated them against a cohort of Indian participants (n = 106). The body mass index (BMI) and waist-to-hip ratio (WHR) were obtained using anthropometric measurements, and body fat mass (BFM), percentage body fat (PBF), and visceral fat area (VFA) were measured using body composition analysis. Facial pictures were also collected and processed to characterize facial geometry. Regression analysis was conducted to determine correlations between body fat parameters and facial model parameters. RESULTS: Lower facial geometry was highly correlated with BMI (R 2 = 0.77) followed by PBF (R 2 = 0.72), VFA (R 2 = 0.65), WHR (R 2 = 0.60), BFM (R 2 = 0.59), and weight (R 2 = 0.54). CONCLUSIONS: The ability to predict obesity using facial images through mobile application or telemedicine can help with early diagnosis and timely medical intervention for people with obesity during the pandemic.