Face recognition based on spatial and temporal proximity

Abstract

In one embodiment, a method includes accessing an image file associated with a first user of a communication system and detecting a face in an image corresponding to the image file. The method also includes accessing an event database associated with the communication system, the event database containing one or more events, each being associated with the first user and one or more second users of the communication system. The method also includes determining one or more candidates among the second users to be matched to the face, where each candidate is associated with an event in the communication system, and where a time associated with the image is in temporal proximity to a time associated with the event.

What is claimed is: 
     
         1 . A method comprising:
 by one or more computing devices, accessing an image file associated with a first user of a communication system;   by the one or more computing devices, detecting one or more faces in an image corresponding to the image file;   by the one or more computing devices, matching one or more of the detected faces to one or more second users associated with the communication system;   by the one or more computing devices, retrieving profile information, social-graph information, or affiliation information associated with one or more of the second users from a user profile database associated with the communication system; and   by the one or more computing devices, for at least one of the matched faces, providing for display a frame in proximity to the matched face, wherein the frame comprises the retrieved information associated with the second user matched to the face, and wherein the frame is at least partially overlaying the image.   
     
     
         2 . The method of  claim 1 , wherein the frame further comprises one or more selectable icons. 
     
     
         3 . The method of  claim 2 , further comprising:
 by the one or more computing devices, receiving a user input associated with a selection of one of the selectable icons; and   by the one or more computing devices, tagging the second user matched to the face, wherein the second user is associated with the received user input.   
     
     
         4 . The method of  claim 2 , further comprising:
 by the one or more computing devices, receiving a user input associated with a selection of one of the selectable icons; and   by the one or more computing devices, sending a friend invitation to the second user matched to the face, wherein the second user is associated with the received user input, and wherein the friend invitation is from the first user.   
     
     
         5 . The method of  claim 2 , further comprising:
 by the one or more computing devices, receiving a user input associated with a selection of one of the selectable icons;   by the one or more computing devices, storing the image file in a media database in response to the received user input; and   by the one or more computing devices, publishing the image file to the first user's profile page.   
     
     
         6 . The method of  claim 2 , wherein the one or more selectable icons are a text string, a picture or a graphical icon. 
     
     
         7 . The method of  claim 1 , wherein the image file comprises:
 a photograph and associated metadata in digital form; or   a video and associated metadata in digital form.   
     
     
         8 . The method of  claim 1 , wherein the matching one or more of the detected faces to the one or more second users is at least based on:
 a comparison of temporal and spatial proximity of the second users, as indicated by a social-networking system, to a time and place associated with the image with a threshold temporal and spatial proximity; and   a comparison of degrees of separation between the second users and the first user indicated by the social-networking system with a threshold degrees of separation.   
     
     
         9 . The method of  claim 8 , wherein the matching one or more of the detected faces to the one or more second users is further based on an application of a facial-recognition algorithm to each of the faces. 
     
     
         10 . The method of  claim 1 , further comprising, by the one or more computing devices, providing for display a box for each of the detected faces, wherein the box encloses each of the detected faces. 
     
     
         11 . A system comprising: one or more processors; and a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to:
 access an image file associated with a first user of a communication system;   detect one or more faces in an image corresponding to the image file;   match one or more of the detected faces to one or more second users associated with the communication system;   retrieve profile information, social-graph information, or affiliation information associated with one or more of the second users from a user profile database associated with the communication system; and   provide for display, for at least one of the matched faces, a frame in proximity to the matched face, wherein the frame comprises the retrieved information associated with the second user matched to the face, and wherein the frame is at least partially overlaying the image.   
     
     
         12 . The system of  claim 11 , wherein the frame further comprises one or more selectable icons. 
     
     
         13 . The system of  claim 12 , wherein the processors are further operable when executing the instructions to:
 receive a user input associated with a selection of one of the selectable icons; and   tag the second user matched to the face, wherein the second user is associated with the received user input.   
     
     
         14 . The system of  claim 12 , wherein the processors are further operable when executing the instructions to:
 receive a user input associated with a selection of one of the selectable icons; and   send a friend invitation to the second user matched to the face, wherein the second user is associated with the received user input, and wherein the friend invitation is from the first user.   
     
     
         15 . The system of  claim 12 , wherein the processors are further operable when executing the instructions to:
 receive a user input associated with a selection of one of the selectable icons;   store the image file in a media database in response to the received user input; and   publish the image file to the first user's profile page.   
     
     
         16 . The system of  claim 12 , wherein the one or more selectable icons are a text string, a picture or a graphical icon. 
     
     
         17 . The system of  claim 11 , wherein the image file comprises:
 a photograph and associated metadata in digital form; or   a video and associated metadata in digital form.   
     
     
         18 . The system of  claim 11 , wherein the matching one or more of the detected faces to the one or more second users is at least based on:
 a comparison of temporal and spatial proximity of the second users, as indicated by a social-networking system, to a time and place associated with the image with a threshold temporal and spatial proximity; and   a comparison of degrees of separation between the second users and the first user indicated by the social-networking system with a threshold degrees of separation.   
     
     
         19 . The system of  claim 18 , wherein the matching one or more of the detected faces to the one or more second users is further based on an application of a facial-recognition algorithm to each of the faces. 
     
     
         20 . One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
 access an image file associated with a first user of a communication system;   detect one or more faces in an image corresponding to the image file;   match one or more of the detected faces to one or more second users associated with the communication system;   retrieve profile information, social-graph information, or affiliation information associated with one or more of the second users from a user profile database associated with the communication system; and   provide for display, for at least one of the matched faces, a frame in proximity to the matched face, wherein the frame comprises the retrieved information associated with the second user matched to the face, and wherein the frame is at least partially overlaying the image.