Using Facial Attractiveness as a Soft Biometric Trait to Enhance Face Recognition Performance

Soft biometrics are traits or features providing information about an individual but cannot be used for individual authentication due to their lack of distinctiveness and permanence. They can enhance human-based identification systems, including face recognition. To optimize face recognition using soft biometrics, facial features can be derived manually (human labeling), automatically, or semiautomatically. Several features are typically associated with soft biometrics, including age, gender, ethnicity, eye color, and weight. Additional features pertinent to face recognition include smiling, expression, wearing glasses, and facial scars or tattoos. One facial feature not yet considered for identification is attractiveness based on facial characteristics. Attractiveness has been found to play a significant role in recognition, similar to gender. This study investigates the relationship between attractiveness and beauty and their implications in recognition, psychology, philosophy, and automated analysis. Previous approaches focusing on attractiveness were not well-formulated for recognition purposes. This chapter introduces a novel approach that incorporates attractiveness as a facial feature. This feature is derived by comparing faces and then ranking the comparisons, making it descriptive in a biometric sense, and it is demonstrated here that attractiveness can indeed aid the recognition process. Our results show that facial attractiveness improves face recognition performance by more than 3% in a soft biometric framework and by 4% in attractiveness-guided automatic recognition systems, in the LFW dataset. These results signify substantial improvements achieved by incorporating attractiveness in two different face recognition frameworks, using standard baseline approaches rather than deep learning to fully establish the fundamental nature of this newly proposed feature.