Facial Recognition Detects Your Friends in Facebook Photos
Another recent publication featured face recognition technology and a proposal for a new type of collaborative face recognition framework that would strive to improve face annotation accuracy by using a combination of face recognition engines available on online social networks.
The research was published in a journal called IEEE Transactions on Multimedia, and is made up of two major parts, the selection of facial recognition engines and the merging of multiple facial recognition results. The collection of face recognition engines hopes to determine the personalized engines suitable for identifying query face images belonging to a particular member of an online social network.
The study takes advantage of various multiple facial recognition results and uses a traditional manner for combining classifier results. Typically, online social networks like Facebook and Myspace are used for sharing or managing a photo or video collection. Users tag photos of individuals and friends using face annotation. The thing is, most online social networks feature manual face annotation, a time-consuming task that can be extremely labor-intensive. The amount of photos posted online increased astronomically each day and that’s a lot of photos and faces to tag. The paper proposes eliminating the need for manual face tagging, presenting an automatic face annotation system based on face detection algorithms.
Face detection algorithms have improved greatly in recent years but they still suffer from a relatively low accuracy when they deal with personal photos, low lighting, variation in angle and pose, and spatial resolution.
For the research, the author created an experimental study to investigate the effectiveness and efficiency of a collaborative facial recognition framework. Using 547,991 photos from Cyworld, an online social network based in Korea, the authors collected personal photos to construct a series of test beds.
Overall they found that the amount of attention that is paid to the creation of decentralized online social networks increases, and is able to eventually overcome issues related to privacy, security, and data ownership.
No related articles.