Abstract: Auto face annotation is important in many real world knowledge management system and multimedia information. Face annotation is a field of face detection and recognition. Mining weakly labeled web facial images on the internet has emerged as a promising paradigm towards auto face annotation. The System examines the problem of celebrity face naming in unconstrained video with user provided metadata. Normally we depend on accurate face labels for supervised learning. But sometimes the faces were not properly annotated. One of the solution used in the proposed system is that; it uses the two parameters a rich set of relationships automatically derived from video content and knowledge from image domain. Relationship is the appearance of faces under different context and their visual similarities. The knowledge includes Web images weakly tagged with celebrity names and the celebrity social networks. The relationships and knowledge is elegantly encoded using conditional random field (CRF) for label inference. Two versions of face annotation are considered: within-video and between-video face labeling. The system further rectifies the error in the metadata to correct false labels and annotate the faces with missing names in the metadata of a video by considering a group of socially connected videos for joint label inference. The system leads to higher accuracy in face labeling than several existing approaches but with minor degradation in speed efficiency.

Keywords: Celebrity face naming, face annotation, social network, unconstrained web video, weak label.