Abstract :
Face annotation is the process of naming a
person through his photo or person involved in video chat.
Face annotation in images and videos enjoys many
potential applications in multimedia information retrieval.
Face annotation usually requires many training data
labelled by hand in order to build effective classifiers. The
complexity in face annotation is how to effectively perform
annotation by identifying the person face. We propose
Search/Query based face annotation scheme. First we
retrieve a shortlist of Top k most similar images from
weakly labelled image database. Then annotate the image
by mapping image to the similar images that exist in image
database. We use clustering based approximation algorithm
to improve the scalability.
Keyword :
Face annotation, weakly labelled images