| 摘要 |
In particular embodiments, one or more images associated with a primary user are received. The image(s) may comprise single images, a series of related images, or video frames. In each image, one or more faces are detected and/or tracked. For each face, a set of one or more candidates are selected who may be identified with the face. The primary user has a computed measure of affinity for candidates in the set through a social network, or the candidate in the set is otherwise known to the primary user. A facial recognition score is calculated for each candidate. A subset of candidates is selected, wherein each candidate in the subset has a facial recognition score above a predetermined threshold. A candidate score is calculated for each candidate based on the facial recognition score and the computed measure of affinity. A winning candidate is selected based on the candidate scores. |
| 主权项 |
1. A method comprising:
by one or more computing devices, receiving one or more images associated with a primary user, wherein at least one of the one or more images includes one or more detected faces; by the one or more computing devices, for each of the one or more detected faces, selecting a set of one or more candidates, wherein the set of one or more candidates comprises:
users of a social network wherein each of the users are associated with a computed measure of affinity of the primary user for each candidate through the social network that exceeds a first predetermined threshold; andusers determined by the social network to be public figures, wherein the public figures are each users of the social network that do not have a computed measure of affinity of the primary user for each candidate through the social network that exceeds the first predetermined threshold; by the one or more computing devices, for each of the one or more detected faces, selecting a subset of one or more candidates from the set of candidates based on a facial recognition score calculating by a facial recognition comparison of each candidate to the detected face, wherein each of the candidates in the subset of candidates has a facial recognition score above a second predetermined threshold; by the one or more computing devices, for each of the detected faces, calculating a candidate score for each of the candidates, wherein the calculation is based on:
the facial recognition score for the candidate; andthe computed measure of affinity for the candidate; and by the one or more computing devices, for each of the one or more faces, selecting a winning candidate from the subset of candidates to identify the face based on the candidate scores for the face. |
