How many times did you saw a person you think you have seen before and not able to recall ? How many times did you wish you knew that blonde girl heading your way? Seems like all of the wishes are about to become true with Google’s new patent for social faces. We are very nearly there with automatic face-recognition technology and social media aggregation.
Google Goggles is a Visual search engine for mobile phones which allows users to take a picture of anything, such as a landmark, or a celebrity and then run a search for what/who they are.
Google wishes to expands its plans further when they acquired like.com, a visual search engine. Prior to Google’s acquisition, Like.com helped online shoppers search site called Riya which used facial recognition technology to search images that users had uploaded and tagged. The latest patent filed by Google: User Interface for Presenting Search Results for Multiple Regions of a Visual Query suggests that Google’s visual search for smartphones might be evolving into something that’s totally new and powerful.
Another Google patent, Facial Recognition With Social Network Aiding, describes facial recognition searches for “one or more likely names” of the people in the image based upon “communications applications, social networking applications, calendar applications, and collaborative applications” to create a list of possible identities for each person. The patent description also mentions that if the photo is tagged with a person’s name, “that picture might be used in future facial recognition queries to recognize the person.”
After receiving the visual query with one or more facial images, the system identifies images that potentially match the respective facial image in accordance with visual similarity criteria. Then one or more persons associated with the potential images are identified. For each identified person, person-specific data comprising metrics of social connectivity to the requester are retrieved from a plurality of applications such as communications applications, social networking applications, calendar applications, and collaborative applications. An ordered list of persons is then generated by ranking the identified persons in accordance with at least metrics of visual similarity between the respective facial image and the potential image matches and with the social connection metrics. Finally, at least one person identifier from the list is sent to the requester.
Benefits aside, Google could face serious privacy allegations. In regards to privacy, Google has several possible scenarios: to send only one of the “identifiers” to the person searching; to possibly allow only the person identified to make the photo public, or to send a request after a person is positively identified, asking if the image can be a face search result for other people’s visual queries from within their social network, which is kind of tricky as Google would need partnership with Twitter & Facebook for this to work. Its not hard to get the former but latter is what we doubt.
Okay, now that we know Google is serious about social facial recognition, how can we confirm is targetting mobile platforms like Android & iPhone? Coincidently, “David Petrou, Andrew Rabinovich, and Adam Hartwig” who worked on Google Goggles, are the patent filers.
What would make it better is that we would see some competition as well. Viewdle is also working on mobile facial recognition that works on the fly, and might help “government stalkers.” by allowing users to tag and save “faceprints” of people, and “then share the faceprint of them with other Viewdle users so they can recognize that person too!” The photosharing privacy settings are integrated with Facebook privacy settings.
Viewdle - Photo and Video Face Tagging from Viewdle on Vimeo.