Survey on Tag Refinement and Tag Completion for Effective Image Retrieval

Online sharing of images is increasingly becoming popular, resulting in the availability of vast collections of user contributed images that have been annotated with user supplied tags. Many search engines are based on tag matching as tag-based image retrieval (TBIR) is not only efficient but also effective. The performance of TBIR is highly dependent on the quality of user supplied tags. Since many users are not interested in choosing appropriate tags, they are usually incomplete and insufficient to describe the whole semantic content of corresponding images. This degrades the performance TBIR. This paper is a study on various techniques which are used to complete the missing tags and correct the noisy tags for given images thereby improving the retrieval performance.