Towards an ontology based framework for searching multimedia contents on the web

We live in a world where there are huge number of consumers and producers of multimedia content. In this sea of information, finding the right content is like finding a needle in a haystack. Rich annotation of multimedia content during its initial upload on the Web, and further various methodologies for framing search query can be helpful to the user in this regard. In addition to annotation of multimedia content based on the user-provided description, various approaches for annotation and indexing of multimedia files based upon the embedded contents have been presented in the literature. However, annotating multimedia files by using multiple possible sources simultaneously to generate better annotation needs further exploration. We have proposed a framework utilizing these multiple sources of information like text, audio, image, etc. This framework generates annotation based on the contents of user entered description, embedded audio, image analysis, optical character recognition and finally by gathering more information from the Web. This framework provides multiple options to search for content like search by image, audio, video, face and also provides an improved textual search. A system has been implemented based on the proposed framework and the work has also been evaluated.