Finding Celebrity in Web Videos Using Audiovisual Recognition

There are number of video clips available online is upward at a stunningstride. Conservatively, user-supplied metadata text, such as the title of the video and a set of keywords, has been the only source of indexing information for user-uploaded videos. Automated extraction of video content for unconstrained and large scale video databases is a challenging and yet baffling problem. In this paper, we current an audiovisual celebrity recognition system nearreflex tagging of unconstrained web videos. Earlier work on audio-visual person recognition depend on the fact that the person in the video is speaking and the structures extracted from audio and visual domain are allied with each other throughout the video. However, this assumption is not valid on unconstrained web videos. Projected method finds the audiovisual mapping and hence improves upon the association assumption. Considering the scale of the application, all pieces of the system are trained automatically without any human supervision. We present the results on 26,000 videos and show the effectiveness of the method per-celebrity basis.