Web Image Re-Ranking using Query Specific Semantic Signatures in a Search Engine
Main Article Content
Abstract
The implementation of Image re-ranking, as a useful way towards progress the grades of web-based image explore, has be adopted with existing trade exploration engines such as Bing and Google. Agreed a query keyword, a collection of picture is first retrieved based on documentary information. Through asking the user to choose a query image from the group, the enduring images are re-ranked based on their visual similarities with the query picture. A main test is that the similarities of visual description do not fine associate with images’ semantic meanings which understand users’ search meaning. Newly people future to competition images in a semantic space which used attributes or reference classes intimately related to the semantic meanings of images as source. However, knowledge a universal visual semantic space to characterize extremely assorted images from the web is tricky and incompetent. In this paper, we suggest a original image re-ranking framework, which automatically offline learns changed semantic spaces for different query keywords. The visual features of images are planned into their connected semantic spaces to get semantic signatures. At the online period, images are re-ranked by comparing their semantic signatures obtained from the semantic space particular by the query keyword. The planned query-specific semantic signatures considerably recover both the precision and competence of image re-ranking. The unique visual features of thousands of scope can be projected to the semantic signatures as short as 25 dimensions. Tentative results show that 25-40 percent relative improvement has been achieved on re-ranking precisions compared with the high-tech methods.
Keywords: Image re-ranking, Image processing, image recovery, semantic, SVM
Downloads
Article Details
COPYRIGHT
Submission of a manuscript implies: that the work described has not been published before, that it is not under consideration for publication elsewhere; that if and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the publisher.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
- The journal allows the author(s) to retain publishing rights without restrictions.
- The journal allows the author(s) to hold the copyright without restrictions.