Web Image Re-ranking Using Semantic Signatures
- Raksha.R.Urs
- H.Suma
- Chitra.R
Image retrieval, semantic signatures, query-specific
Image re-ranking is one of the most effective technique to improve the results of web-based image search. This technique is adopted by current commercial search engines such as Bing & Google. In this paper we propose a novel image re-ranking framework with two stages , offline and online stage. In the offline stage, different semantic spaces for different query keywords are automatically learned and the visual features of images are projected into their related semantic spaces to get semantic signatures. At the online stage, images are re-ranked by comparing their semantic signatures obtained from the semantic space specified by the query keyword. The proposed system significantly improves both the accuracy and efficiency of image re-ranking by making use of query-specific semantic signatures.
Raksha.R.Urs, H.Suma, Chitra.R. "Web Image Re-ranking Using Semantic Signatures".INTERNATIONAL JOURNAL OF ENGINEERING DEVELOPMENT AND RESEARCH ISSN:2321-9939, Vol.3, Issue 2, pp.345-348, URL :https://rjwave.org/ijedr/papers/IJEDR1502065.pdf
Volume 3 Issue 2
Pages. 345-348