LFW (Labeled Faces in the Wild) dataset is a face photo database developed to explore the problem of unrestricted face recognition. LFW was released for research purposes to make advancements on face verification, not to conduct a comprehensive review of commercial algorithms prior to release. The database is an initial attempt to provide a set of categorized faces covering a range of circumstances that people commonly encounter in their daily lives. The database displays "normal" diversity in poses, lighting, focus, accuracy, facial expressions, age, gender, ethnicity, accessories, makeup, occlusions, background, and photographic quality. Despite this discrepancy, the images in the database are presented in a simple and consistent format for maximum usability.