Towards understanding the character of quality sampling in deep learning face recognition

Abstract

Face recognition has become one of the most important modalities of biometrics in recent years. It widely utilises deep learning computer vision tools and adopts large collections of unconstrained face images of celebrities for training. Such choice of the data is related to its public availability when existing document compliant face image collections are hardly accessible due to security and privacy issues. Such inconsistency between the training data and deploy scenario may lead to a leak in performance in biometric systems, which are developed specifically for dealing with ID document compliant images. To mitigate this problem, we propose to regularise the training of the deep face recognition network with a specific sample mining strategy, which penalises the samples by their estimated quality. In addition to several considered quality metrics in recent work, we also expand our deep learning strategy to other sophisticated quality estimation methods and perform experiments to better understand the nature of quality sampling. Namely, we seek for the penalising manner (sampling character) that better satisfies the purpose of adapting deep learning face recognition for images of ID and travel documents. Extensive experiments demonstrate the efficiency of the approach for ID document compliant face images.