Described herein are systems and methods that may autonomously identify a person of a member pool based on pictures of the person, without requiring the person's cooperation. Contextual information of the picture(s) along with the picture(s) are utilized. Contextual information may be the information that is related to current circumstances when the picture was taken. A system may comprise cameras that map appearances to visual data; a mapping function that maps identities and camera information to generate contextual information and a set of a priori probabilities; recognition functions that map the visual data to another set of probabilities, which match each of the visual data with each of the one of the identities; and a decision function that combines the set of a priori probabilities and the another set of probabilities to determine one of the plurality of the identities.