PROJECT TITLE:

Tag Based Image Search by Social Re-ranking - 2016

ABSTRACT:

Social media sharing websites like Flickr allow users to annotate images with free tags, that significantly contribute to the development of the internet Image Retrieval and organization. Tag-based image search is a vital methodology to find images contributed by social users in such social websites. However, how to create the top ranked result relevant and, with diversity, is challenging. In this paper, we tend to propose a social re-ranking system for tag-based image retrieval with the thought of a picture's relevance and diversity. We have a tendency to aim at re-ranking pictures in keeping with their visual data, semantic info, and social clues. The initial results embody images contributed by totally different social users. Usually every user contributes several pictures. First, we type these pictures by inter-user re-ranking. Users that have higher contribution to the given query rank higher. Then we have a tendency to sequentially implement intra-user re-ranking on the ranked user's image set, and only the most relevant image from every user's image set is selected. These selected pictures compose the ultimate retrieved results. We have a tendency to build an inverted index structure for the social image dataset to accelerate the looking out process. Experimental results on a Flickr dataset show that our social re-ranking methodology is effective and efficient.