A Survey on Multimedia Search Optimization based on Multimodal Information Resources

This survey constitutes a literature study that overviews the state-of-the-art in multimedia search. Techniques that are covered include multimodal re-ranking, pseudo-relevance feedback, query classification and query suggestion. Discussion of the techniques will be related to the various relevance criteria that have been addressed in the literature. The survey will set particular focus on techniques that have been deployed in social communities, where copious amounts of auxiliary information in the form of social enrichments are available to improve search.